/[pcre]/code/trunk/ChangeLog
ViewVC logotype

Diff of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 25 by nigel, Sat Feb 24 21:38:45 2007 UTC revision 388 by ph10, Wed Mar 11 17:03:17 2009 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.9 xx-xxx-09
5    ---------------------
6    
7    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
8        (pcretest), all targets were linked against these libraries. This included
9        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
10        libraries. This caused unwanted dependencies to be created. This problem
11        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
12        pcretest is linked with readline.
13    
14    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
15        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
16        moved outside it again, because FALSE and TRUE are already defined in AIX,
17        but BOOL is not.
18    
19    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
20        PCRE_MATCH_LIMIT_RETURSION values as ints, when they should be long ints.
21    
22    4.  The pcregrep documentation said spaces were inserted as well as colons (or
23        hyphens) following file names and line numbers when outputting matching
24        lines. This is not true; no spaces are inserted. I have also clarified the
25        wording for the --colour (or --color) option.
26    
27    5.  In pcregrep, when --colour was used with -o, the list of matching strings
28        was not coloured; this is different to GNU grep, so I have changed it to be
29        the same.
30    
31    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
32        each matching line was coloured. Now it goes on to look for further matches
33        of any of the test patterns, which is the same behaviour as GNU grep.
34    
35    7.  A pattern that could match an empty string could cause pcregrep to loop; it
36        doesn't make sense to accept an empty string match in pcregrep, so I have
37        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
38        seems to be how GNU grep behaves.
39    
40    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
41        start or after a newline", because the conditional assertion was not being
42        correctly handled. The rule now is that both the assertion and what follows
43        in the first alternative must satisfy the test.
44    
45    9.  If auto-callout was enabled in a pattern with a conditional group, PCRE
46        could crash during matching.
47    
48    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
49        used for matching.
50    
51    11. Unicode property support in character classes was not working for
52        characters (bytes) greater than 127 when not in UTF-8 mode.
53    
54    12. Added the -M command line option to pcretest.
55    
56    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
57    
58    
59    Version 7.8 05-Sep-08
60    ---------------------
61    
62    1.  Replaced UCP searching code with optimized version as implemented for Ad
63        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
64        stage table and inline lookup instead of a function, giving speed ups of 2
65        to 5 times on some simple patterns that I tested. Permission was given to
66        distribute the MultiStage2.py script that generates the tables (it's not in
67        the tarball, but is in the Subversion repository).
68    
69    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
70        scripts.
71    
72    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
73        a group with a zero qualifier. The result of the study could be incorrect,
74        or the function might crash, depending on the pattern.
75    
76    4.  Caseless matching was not working for non-ASCII characters in back
77        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
78        It now works when Unicode Property Support is available.
79    
80    5.  In pcretest, an escape such as \x{de} in the data was always generating
81        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
82        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
83        truncation.
84    
85    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
86    
87    7.  Added two (int) casts to pcregrep when printing the difference of two
88        pointers, in case they are 64-bit values.
89    
90    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
91        test 2 if it fails.
92    
93    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
94        and a #define of that name to empty if it is not externally set. This is to
95        allow users of MSVC to set it if necessary.
96    
97    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
98        the convenience functions in the pcre_get.c source file.
99    
100    11. An option change at the start of a pattern that had top-level alternatives
101        could cause overwriting and/or a crash. This command provoked a crash in
102        some environments:
103    
104          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
105    
106        This potential security problem was recorded as CVE-2008-2371.
107    
108    12. For a pattern where the match had to start at the beginning or immediately
109        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
110        pcre_dfa_exec() could read past the end of the passed subject if there was
111        no match. To help with detecting such bugs (e.g. with valgrind), I modified
112        pcretest so that it places the subject at the end of its malloc-ed buffer.
113    
114    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
115        exec() might read past the end of the data buffer in UTF-8 mode.
116    
117    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
118        the data contained the byte 0x85 as part of a UTF-8 character within its
119        first line. This applied both to normal and DFA matching.
120    
121    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
122        /^[^d]*?$/8 failed to match "abc".
123    
124    16. Added a missing copyright notice to pcrecpp_internal.h.
125    
126    17. Make it more clear in the documentation that values returned from
127        pcre_exec() in ovector are byte offsets, not character counts.
128    
129    18. Tidied a few places to stop certain compilers from issuing warnings.
130    
131    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
132        supplied by Stefan Weber. I made a further small update for 7.8 because
133        there is a change of source arrangements: the pcre_searchfuncs.c module is
134        replaced by pcre_ucd.c.
135    
136    
137    Version 7.7 07-May-08
138    ---------------------
139    
140    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
141        a string to a long long, pretend we don't even have a long long." This is
142        done by checking for the strtoq, strtoll, and _strtoi64 functions.
143    
144    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
145        pre-7.6 versions, which defined a global no_arg variable instead of putting
146        it in the RE class. (See also #8 below.)
147    
148    3.  Remove a line of dead code, identified by coverity and reported by Nuno
149        Lopes.
150    
151    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
152    
153        (1) The include/exclude patterns were being applied to the whole pathnames
154            of files, instead of just to the final components.
155    
156        (2) If there was more than one level of directory, the subdirectories were
157            skipped unless they satisfied the include/exclude conditions. This is
158            inconsistent with GNU grep (and could even be seen as contrary to the
159            pcregrep specification - which I improved to make it absolutely clear).
160            The action now is always to scan all levels of directory, and just
161            apply the include/exclude patterns to regular files.
162    
163    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
164        --exclude_dir in the tests to avoid scanning .svn directories.
165    
166    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
167        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
168        doesn't support NULs in patterns.
169    
170    7.  Added some missing "const"s to declarations of static tables in
171        pcre_compile.c and pcre_dfa_exec.c.
172    
173    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
174        caused by fix #2  above. (Subsequently also a second patch to fix the
175        first patch. And a third patch - this was a messy problem.)
176    
177    9.  Applied Craig's patch to remove the use of push_back().
178    
179    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
180        matching function regexec().
181    
182    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
183        which, however, unlike Perl's \g{...}, are subroutine calls, not back
184        references. PCRE supports relative numbers with this syntax (I don't think
185        Oniguruma does).
186    
187    12. Previously, a group with a zero repeat such as (...){0} was completely
188        omitted from the compiled regex. However, this means that if the group
189        was called as a subroutine from elsewhere in the pattern, things went wrong
190        (an internal error was given). Such groups are now left in the compiled
191        pattern, with a new opcode that causes them to be skipped at execution
192        time.
193    
194    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
195        to the way PCRE behaves:
196    
197        (a) A lone ] character is dis-allowed (Perl treats it as data).
198    
199        (b) A back reference to an unmatched subpattern matches an empty string
200            (Perl fails the current match path).
201    
202        (c) A data ] in a character class must be notated as \] because if the
203            first data character in a class is ], it defines an empty class. (In
204            Perl it is not possible to have an empty class.) The empty class []
205            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
206            The negative empty class [^] matches any one character, independently
207            of the DOTALL setting.
208    
209    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
210        non-existent subpattern following a character class starting with ']' and
211        containing () gave an internal compiling error instead of "reference to
212        non-existent subpattern". Fortunately, when the pattern did exist, the
213        compiled code was correct. (When scanning forwards to check for the
214        existencd of the subpattern, it was treating the data ']' as terminating
215        the class, so got the count wrong. When actually compiling, the reference
216        was subsequently set up correctly.)
217    
218    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
219        it was being rejected as not supported by pcre_dfa_exec(), even though
220        other assertions are supported. I have made pcre_dfa_exec() support
221        (*FAIL).
222    
223    16. The implementation of 13c above involved the invention of a new opcode,
224        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
225        cannot be changed at match time, I realized I could make a small
226        improvement to matching performance by compiling OP_ALLANY instead of
227        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
228        on the OP_ANY path.
229    
230    17. Compiling pcretest on Windows with readline support failed without the
231        following two fixes: (1) Make the unistd.h include conditional on
232        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
233    
234    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
235        ncurses library to be included for pcretest when ReadLine support is
236        requested, but also to allow for it to be overridden. This patch came from
237        Daniel Bergström.
238    
239    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
240        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
241        any errors with the current Unicode tables. Thanks to Peter Kankowski for
242        spotting this.
243    
244    
245    Version 7.6 28-Jan-08
246    ---------------------
247    
248    1.  A character class containing a very large number of characters with
249        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
250        overflow.
251    
252    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
253        HAVE_LONG_LONG is not defined.
254    
255    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
256        bring them up to date and include new features. This patch includes:
257    
258        - Fixed PH's badly added libz and libbz2 support.
259        - Fixed a problem with static linking.
260        - Added pcredemo. [But later removed - see 7 below.]
261        - Fixed dftables problem and added an option.
262        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
263            HAVE_LONG_LONG.
264        - Added readline support for pcretest.
265        - Added an listing of the option settings after cmake has run.
266    
267    4.  A user submitted a patch to Makefile that makes it easy to create
268        "pcre.dll" under mingw when using Configure/Make. I added stuff to
269        Makefile.am that cause it to include this special target, without
270        affecting anything else. Note that the same mingw target plus all
271        the other distribution libraries and programs are now supported
272        when configuring with CMake (see 6 below) instead of with
273        Configure/Make.
274    
275    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
276        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
277        exported in the Windows port". It has not yet been confirmed that the patch
278        solves the problem, but it does no harm.
279    
280    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
281        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
282        with CMake, and also correct the comment about stack recursion.
283    
284    7.  Remove the automatic building of pcredemo from the ./configure system and
285        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
286        of a program that users should build themselves after PCRE is installed, so
287        building it automatically is not really right. What is more, it gave
288        trouble in some build environments.
289    
290    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
291    
292    
293    Version 7.5 10-Jan-08
294    ---------------------
295    
296    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
297        values in parens when parsing an RE using the C++ wrapper."
298    
299    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
300        Characters greater than 255 were excluded from the class instead of being
301        included.
302    
303    3.  The same bug as (2) above applied to negated POSIX classes such as
304        [:^space:].
305    
306    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
307        defined or documented. It seems to have been a typo for PCRE_STATIC, so
308        I have changed it.
309    
310    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
311        first named subpattern) and a construct such as (?&a) would reference the
312        first named subpattern whose name started with "a" (in other words, the
313        length check was missing). Both these problems are fixed. "Subpattern name
314        expected" is now given for (?&) (a zero-length name), and this patch also
315        makes it give the same error for \k'' (previously it complained that that
316        was a reference to a non-existent subpattern).
317    
318    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
319        this is right because (?- can be followed by option settings as well as by
320        digits. I have, however, made the messages clearer.
321    
322    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
323        than the number used in the conditional) now cause a compile-time error.
324        This is actually not compatible with Perl, which accepts such patterns, but
325        treats the conditional as always being FALSE (as PCRE used to), but it
326        seems to me that giving a diagnostic is better.
327    
328    8.  Change "alphameric" to the more common word "alphanumeric" in comments
329        and messages.
330    
331    9.  Fix two occurrences of "backslash" in comments that should have been
332        "backspace".
333    
334    10. Remove two redundant lines of code that can never be obeyed (their function
335        was moved elsewhere).
336    
337    11. The program that makes PCRE's Unicode character property table had a bug
338        which caused it to generate incorrect table entries for sequences of
339        characters that have the same character type, but are in different scripts.
340        It amalgamated them into a single range, with the script of the first of
341        them. In other words, some characters were in the wrong script. There were
342        thirteen such cases, affecting characters in the following ranges:
343    
344          U+002b0 - U+002c1
345          U+0060c - U+0060d
346          U+0061e - U+00612
347          U+0064b - U+0065e
348          U+0074d - U+0076d
349          U+01800 - U+01805
350          U+01d00 - U+01d77
351          U+01d9b - U+01dbf
352          U+0200b - U+0200f
353          U+030fc - U+030fe
354          U+03260 - U+0327f
355          U+0fb46 - U+0fbb1
356          U+10450 - U+1049d
357    
358    12. The -o option (show only the matching part of a line) for pcregrep was not
359        compatible with GNU grep in that, if there was more than one match in a
360        line, it showed only the first of them. It now behaves in the same way as
361        GNU grep.
362    
363    13. If the -o and -v options were combined for pcregrep, it printed a blank
364        line for every non-matching line. GNU grep prints nothing, and pcregrep now
365        does the same. The return code can be used to tell if there were any
366        non-matching lines.
367    
368    14. Added --file-offsets and --line-offsets to pcregrep.
369    
370    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
371        infinitely looping recursion. The bug was that positive lookaheads were not
372        being skipped when checking for a possible empty match (negative lookaheads
373        and both kinds of lookbehind were skipped).
374    
375    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
376        inclusion of <windows.h> to before rather than after the definition of
377        INVALID_FILE_ATTRIBUTES (patch from David Byron).
378    
379    17. Specifying a possessive quantifier with a specific limit for a Unicode
380        character property caused pcre_compile() to compile bad code, which led at
381        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
382        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
383        caused the error; without that there was no problem.
384    
385    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
386    
387    19. Added --enable-pcretest-libreadline.
388    
389    20. In pcrecpp.cc, the variable 'count' was incremented twice in
390        RE::GlobalReplace(). As a result, the number of replacements returned was
391        double what it should be. I removed one of the increments, but Craig sent a
392        later patch that removed the other one (the right fix) and added unit tests
393        that check the return values (which was not done before).
394    
395    21. Several CMake things:
396    
397        (1) Arranged that, when cmake is used on Unix, the libraries end up with
398            the names libpcre and libpcreposix, not just pcre and pcreposix.
399    
400        (2) The above change means that pcretest and pcregrep are now correctly
401            linked with the newly-built libraries, not previously installed ones.
402    
403        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
404    
405    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
406        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
407        UTF-8 newline character). The key issue is that the pattern starts .*;
408        this means that the match must be either at the beginning, or after a
409        newline. The bug was in the code for advancing after a failed match and
410        checking that the new position followed a newline. It was not taking
411        account of UTF-8 characters correctly.
412    
413    23. PCRE was behaving differently from Perl in the way it recognized POSIX
414        character classes. PCRE was not treating the sequence [:...:] as a
415        character class unless the ... were all letters. Perl, however, seems to
416        allow any characters between [: and :], though of course it rejects as
417        unknown any "names" that contain non-letters, because all the known class
418        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
419        for example, whereas PCRE did not - it did not recognize a POSIX character
420        class. This seemed a bit dangerous, so the code has been changed to be
421        closer to Perl. The behaviour is not identical to Perl, because PCRE will
422        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
423        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
424        Perl does, and where it didn't before.
425    
426    24. Rewrite so as to remove the single use of %n from pcregrep because in some
427        Windows environments %n is disabled by default.
428    
429    
430    Version 7.4 21-Sep-07
431    ---------------------
432    
433    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
434        means that a class such as [\s] counted as "explicit reference to CR or
435        LF". That isn't really right - the whole point of the change was to try to
436        help when there was an actual mention of one of the two characters. So now
437        the change happens only if \r or \n (or a literal CR or LF) character is
438        encountered.
439    
440    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
441        of both had grown to the point where there were only 3 bits left.
442        Fortunately, there was spare space in the data structure, and so I have
443        moved the internal flags into a new 16-bit field to free up more option
444        bits.
445    
446    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
447        but did not set the internal JCHANGED flag - either of these is enough to
448        control the way the "get" function works - but the PCRE_INFO_JCHANGED
449        facility is supposed to tell if (?J) was ever used, so now (?J) at the
450        start sets both bits.
451    
452    4.  Added options (at build time, compile time, exec time) to change \R from
453        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
454    
455    5.  doc/pcresyntax.html was missing from the distribution.
456    
457    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
458        compatibility, even though it is no longer used.
459    
460    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
461        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
462        windows.h file is present (where different names are used). [This was
463        reversed later after testing - see 16 below.]
464    
465    8.  Changed all #include <config.h> to #include "config.h". There were also
466        some further <pcre.h> cases that I changed to "pcre.h".
467    
468    9.  When pcregrep was used with the --colour option, it missed the line ending
469        sequence off the lines that it output.
470    
471    10. It was pointed out to me that arrays of string pointers cause lots of
472        relocations when a shared library is dynamically loaded. A technique of
473        using a single long string with a table of offsets can drastically reduce
474        these. I have refactored PCRE in four places to do this. The result is
475        dramatic:
476    
477          Originally:                          290
478          After changing UCP table:            187
479          After changing error message table:   43
480          After changing table of "verbs"       36
481          After changing table of Posix names   22
482    
483        Thanks to the folks working on Gregex for glib for this insight.
484    
485    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
486        unicode-properties was also set.
487    
488    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
489    
490    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
491        checked only for CRLF.
492    
493    14. Added casts to pcretest.c to avoid compiler warnings.
494    
495    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
496    
497    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
498        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
499        entirely. This removes changes made in 7 above.
500    
501    17. The CMake files have been updated, and there is now more information about
502        building with CMake in the NON-UNIX-USE document.
503    
504    
505    Version 7.3 28-Aug-07
506    ---------------------
507    
508     1. In the rejigging of the build system that eventually resulted in 7.1, the
509        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
510        brackets there is not right, since it causes compilers to look for an
511        installed pcre.h, not the version that is in the source that is being
512        compiled (which of course may be different). I have changed it back to:
513    
514          #include "pcre.h"
515    
516        I have a vague recollection that the change was concerned with compiling in
517        different directories, but in the new build system, that is taken care of
518        by the VPATH setting the Makefile.
519    
520     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
521        when the subject happened to end in the byte 0x85 (e.g. if the last
522        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
523        characters but of course it shouldn't be taken as a newline when it is part
524        of another character. The bug was that, for an unlimited repeat of . in
525        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
526        characters when looking for a newline.
527    
528     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
529    
530     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
531        in debug output.
532    
533     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
534        long printing in the pcrecpp unittest when running under MinGW.
535    
536     6. ESC_K was left out of the EBCDIC table.
537    
538     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
539        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
540        limit also applies to "virtual nesting" when a pattern is recursive, and in
541        this case 1000 isn't so big. I have been able to remove this limit at the
542        expense of backing off one optimization in certain circumstances. Normally,
543        when pcre_exec() would call its internal match() function recursively and
544        immediately return the result unconditionally, it uses a "tail recursion"
545        feature to save stack. However, when a subpattern that can match an empty
546        string has an unlimited repetition quantifier, it no longer makes this
547        optimization. That gives it a stack frame in which to save the data for
548        checking that an empty string has been matched. Previously this was taken
549        from the 1000-entry workspace that had been reserved. So now there is no
550        explicit limit, but more stack is used.
551    
552     8. Applied Daniel's patches to solve problems with the import/export magic
553        syntax that is required for Windows, and which was going wrong for the
554        pcreposix and pcrecpp parts of the library. These were overlooked when this
555        problem was solved for the main library.
556    
557     9. There were some crude static tests to avoid integer overflow when computing
558        the size of patterns that contain repeated groups with explicit upper
559        limits. As the maximum quantifier is 65535, the maximum group length was
560        set at 30,000 so that the product of these two numbers did not overflow a
561        32-bit integer. However, it turns out that people want to use groups that
562        are longer than 30,000 bytes (though not repeat them that many times).
563        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
564        made it possible to implement the integer overflow checks in a much more
565        dynamic way, which I have now done. The artificial limitation on group
566        length has been removed - we now have only the limit on the total length of
567        the compiled pattern, which depends on the LINK_SIZE setting.
568    
569    10. Fixed a bug in the documentation for get/copy named substring when
570        duplicate names are permitted. If none of the named substrings are set, the
571        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
572        empty string.
573    
574    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
575        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
576        because the ] is interpreted as the first data character and the
577        terminating ] is not found. PCRE has been made compatible with Perl in this
578        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
579        cause memory overwriting.
580    
581    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
582        string has been matched (to stop an infinite loop). It was not recognizing
583        a conditional subpattern that could match an empty string if that
584        subpattern was within another subpattern. For example, it looped when
585        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
586        condition was not nested. This bug has been fixed.
587    
588    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
589        past the start of the subject in the presence of bytes with the top bit
590        set, for example "\x8aBCD".
591    
592    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
593        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
594    
595    14. Optimized (?!) to (*FAIL).
596    
597    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
598        This restricts code points to be within the range 0 to 0x10FFFF, excluding
599        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
600        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
601        does: it's just the validity check that is more restrictive.
602    
603    16. Inserted checks for integer overflows during escape sequence (backslash)
604        processing, and also fixed erroneous offset values for syntax errors during
605        backslash processing.
606    
607    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
608        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
609    
610    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
611        caused an overrun.
612    
613    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
614        something other than just ASCII characters) inside a group that had an
615        unlimited repeat caused a loop at compile time (while checking to see
616        whether the group could match an empty string).
617    
618    20. Debugging a pattern containing \p or \P could cause a crash. For example,
619        [\P{Any}] did so. (Error in the code for printing property names.)
620    
621    21. An orphan \E inside a character class could cause a crash.
622    
623    22. A repeated capturing bracket such as (A)? could cause a wild memory
624        reference during compilation.
625    
626    23. There are several functions in pcre_compile() that scan along a compiled
627        expression for various reasons (e.g. to see if it's fixed length for look
628        behind). There were bugs in these functions when a repeated \p or \P was
629        present in the pattern. These operators have additional parameters compared
630        with \d, etc, and these were not being taken into account when moving along
631        the compiled data. Specifically:
632    
633        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
634            length.
635    
636        (b) An item such as \pL+ within a repeated group could cause crashes or
637            loops.
638    
639        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
640            "reference to non-existent subpattern" error.
641    
642        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
643    
644    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
645        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
646    
647    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
648    
649    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
650        character were causing crashes (broken optimization).
651    
652    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
653        \p or \P) caused a compile-time loop.
654    
655    28. More problems have arisen in unanchored patterns when CRLF is a valid line
656        break. For example, the unstudied pattern [\r\n]A does not match the string
657        "\r\nA" because change 7.0/46 below moves the current point on by two
658        characters after failing to match at the start. However, the pattern \nA
659        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
660        the same is true. There doesn't seem any very clean way out of this, but
661        what I have chosen to do makes the common cases work: PCRE now takes note
662        of whether there can be an explicit match for \r or \n anywhere in the
663        pattern, and if so, 7.0/46 no longer applies. As part of this change,
664        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
665        pattern has explicit CR or LF references.
666    
667    29. Added (*CR) etc for changing newline setting at start of pattern.
668    
669    
670    Version 7.2 19-Jun-07
671    ---------------------
672    
673     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
674        which is apparently normally available under Windows.
675    
676     2. Re-jig the pcregrep tests with different newline settings in an attempt
677        to make them independent of the local environment's newline setting.
678    
679     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
680    
681     4. Some of the "internals" tests were previously cut out when the link size
682        was not 2, because the output contained actual offsets. The recent new
683        "Z" feature of pcretest means that these can be cut out, making the tests
684        usable with all link sizes.
685    
686     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
687        stack recursion. This gives a massive performance boost under BSD, but just
688        a small improvement under Linux. However, it saves one field in the frame
689        in all cases.
690    
691     6. Added more features from the forthcoming Perl 5.10:
692    
693        (a) (?-n) (where n is a string of digits) is a relative subroutine or
694            recursion call. It refers to the nth most recently opened parentheses.
695    
696        (b) (?+n) is also a relative subroutine call; it refers to the nth next
697            to be opened parentheses.
698    
699        (c) Conditions that refer to capturing parentheses can be specified
700            relatively, for example, (?(-2)... or (?(+3)...
701    
702        (d) \K resets the start of the current match so that everything before
703            is not part of it.
704    
705        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
706    
707        (f) \g{name} is another synonym - part of Perl 5.10's unification of
708            reference syntax.
709    
710        (g) (?| introduces a group in which the numbering of parentheses in each
711            alternative starts with the same number.
712    
713        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
714    
715     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
716        PCRE_INFO_JCHANGED.
717    
718     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
719        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
720        for detecting groups that can match an empty string.
721    
722     9. A pattern with a very large number of alternatives (more than several
723        hundred) was running out of internal workspace during the pre-compile
724        phase, where pcre_compile() figures out how much memory will be needed. A
725        bit of new cunning has reduced the workspace needed for groups with
726        alternatives. The 1000-alternative test pattern now uses 12 bytes of
727        workspace instead of running out of the 4096 that are available.
728    
729    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
730    
731    11. Applied patch from Google to remove an optimization that didn't quite work.
732        The report of the bug said:
733    
734          pcrecpp::RE("a*").FullMatch("aaa") matches, while
735          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
736          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
737    
738    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
739        it matched the wrong number of bytes.
740    
741    
742    Version 7.1 24-Apr-07
743    ---------------------
744    
745     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
746        that is more "standard", making use of automake and other Autotools. There
747        is some re-arrangement of the files and adjustment of comments consequent
748        on this.
749    
750     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
751        for recursive directory scanning broke on some systems because the files
752        are not scanned in any specific order and on different systems the order
753        was different. A call to "sort" has been inserted into RunGrepTest for the
754        approprate test as a short-term fix. In the longer term there may be an
755        alternative.
756    
757     3. I had an email from Eric Raymond about problems translating some of PCRE's
758        man pages to HTML (despite the fact that I distribute HTML pages, some
759        people do their own conversions for various reasons). The problems
760        concerned the use of low-level troff macros .br and .in. I have therefore
761        removed all such uses from the man pages (some were redundant, some could
762        be replaced by .nf/.fi pairs). The 132html script that I use to generate
763        HTML has been updated to handle .nf/.fi and to complain if it encounters
764        .br or .in.
765    
766     4. Updated comments in configure.ac that get placed in config.h.in and also
767        arranged for config.h to be included in the distribution, with the name
768        config.h.generic, for the benefit of those who have to compile without
769        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
770    
771     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
772        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
773        makevp.bat for the current PCRE, using the additional files
774        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
775    
776     6. A Windows user reported a minor discrepancy with test 2, which turned out
777        to be caused by a trailing space on an input line that had got lost in his
778        copy. The trailing space was an accident, so I've just removed it.
779    
780     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
781        that is needed.
782    
783     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
784        as "const" (a) because they are and (b) because it helps the PHP
785        maintainers who have recently made a script to detect big data structures
786        in the php code that should be moved to the .rodata section. I remembered
787        to update Builducptable as well, so it won't revert if ucptable.h is ever
788        re-created.
789    
790     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
791        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
792        order to be able to cut out the UTF-8 tables in the latter when UTF-8
793        support is not required. This saves 1.5-2K of code, which is important in
794        some applications.
795    
796        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
797        so as not to refer to the tables, even though these functions will never be
798        called when UTF-8 support is disabled. Otherwise there are problems with a
799        shared library.
800    
801    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
802    
803        (a) It was defining its arguments as char * instead of void *.
804    
805        (b) It was assuming that all moves were upwards in memory; this was true
806            a long time ago when I wrote it, but is no longer the case.
807    
808        The emulated memove() is provided for those environments that have neither
809        memmove() nor bcopy(). I didn't think anyone used it these days, but that
810        is clearly not the case, as these two bugs were recently reported.
811    
812    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
813        and Detrail to create the HTML documentation, the .txt form of the man
814        pages, and it removes trailing spaces from listed files. It also creates
815        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
816        case, it wraps all the #defines with #ifndefs. This script should be run
817        before "make dist".
818    
819    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
820        with Unicode property support.
821    
822        (a) For a maximizing quantifier, if the two different cases of the
823            character were of different lengths in their UTF-8 codings (there are
824            some cases like this - I found 11), and the matching function had to
825            back up over a mixture of the two cases, it incorrectly assumed they
826            were both the same length.
827    
828        (b) When PCRE was configured to use the heap rather than the stack for
829            recursion during matching, it was not correctly preserving the data for
830            the other case of a UTF-8 character when checking ahead for a match
831            while processing a minimizing repeat. If the check also involved
832            matching a wide character, but failed, corruption could cause an
833            erroneous result when trying to check for a repeat of the original
834            character.
835    
836    13. Some tidying changes to the testing mechanism:
837    
838        (a) The RunTest script now detects the internal link size and whether there
839            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
840            values substituted by "configure". (The RunGrepTest script already did
841            this for UTF-8.) The configure.ac script no longer substitutes the
842            relevant variables.
843    
844        (b) The debugging options /B and /D in pcretest show the compiled bytecode
845            with length and offset values. This means that the output is different
846            for different internal link sizes. Test 2 is skipped for link sizes
847            other than 2 because of this, bypassing the problem. Unfortunately,
848            there was also a test in test 3 (the locale tests) that used /B and
849            failed for link sizes other than 2. Rather than cut the whole test out,
850            I have added a new /Z option to pcretest that replaces the length and
851            offset values with spaces. This is now used to make test 3 independent
852            of link size. (Test 2 will be tidied up later.)
853    
854    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
855        segmentation fault instead of returning the appropriate error message.
856    
857    15. In multiline mode when the newline sequence was set to "any", the pattern
858        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
859        This doesn't seem right; it now treats the CRLF combination as the line
860        ending, and so does not match in that case. It's only a pattern such as ^$
861        that would hit this one: something like ^ABC$ would have failed after \r
862        and then tried again after \r\n.
863    
864    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
865        in an attempt to make files that differ only in their line terminators
866        compare equal. This works on Linux.
867    
868    17. Under certain error circumstances pcregrep might try to free random memory
869        as it exited. This is now fixed, thanks to valgrind.
870    
871    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
872        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
873        was because its rules for how to advance for /g after matching an empty
874        string at the end of a line did not allow for this case. They now check for
875        it specially.
876    
877    20. pcretest is supposed to handle patterns and data of any length, by
878        extending its buffers when necessary. It was getting this wrong when the
879        buffer for a data line had to be extended.
880    
881    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
882        CRLF as a newline sequence.
883    
884    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
885        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
886        I have nevertheless tidied it up.
887    
888    23. Added some casts to kill warnings from HP-UX ia64 compiler.
889    
890    24. Added a man page for pcre-config.
891    
892    
893    Version 7.0 19-Dec-06
894    ---------------------
895    
896     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
897        moving to gcc 4.1.1.
898    
899     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
900        sys/time.h, which is documented as needed for this function. It doesn't
901        seem to matter on Linux, but it showed up on some releases of OS X.
902    
903     3. It seems that there are systems where bytes whose values are greater than
904        127 match isprint() in the "C" locale. The "C" locale should be the
905        default when a C program starts up. In most systems, only ASCII printing
906        characters match isprint(). This difference caused the output from pcretest
907        to vary, making some of the tests fail. I have changed pcretest so that:
908    
909        (a) When it is outputting text in the compiled version of a pattern, bytes
910            other than 32-126 are always shown as hex escapes.
911    
912        (b) When it is outputting text that is a matched part of a subject string,
913            it does the same, unless a different locale has been set for the match
914            (using the /L modifier). In this case, it uses isprint() to decide.
915    
916     4. Fixed a major bug that caused incorrect computation of the amount of memory
917        required for a compiled pattern when options that changed within the
918        pattern affected the logic of the preliminary scan that determines the
919        length. The relevant options are -x, and -i in UTF-8 mode. The result was
920        that the computed length was too small. The symptoms of this bug were
921        either the PCRE error "internal error: code overflow" from pcre_compile(),
922        or a glibc crash with a message such as "pcretest: free(): invalid next
923        size (fast)". Examples of patterns that provoked this bug (shown in
924        pcretest format) are:
925    
926          /(?-x: )/x
927          /(?x)(?-x: \s*#\s*)/
928          /((?i)[\x{c0}])/8
929          /(?i:[\x{c0}])/8
930    
931        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
932        is now done differently.
933    
934     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
935        wrapper classes; (b) implement a new function in the C++ scanner that is
936        more efficient than the old way of doing things because it avoids levels of
937        recursion in the regex matching; (c) add a paragraph to the documentation
938        for the FullMatch() function.
939    
940     6. The escape sequence \n was being treated as whatever was defined as
941        "newline". Not only was this contrary to the documentation, which states
942        that \n is character 10 (hex 0A), but it also went horribly wrong when
943        "newline" was defined as CRLF. This has been fixed.
944    
945     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
946        was being set to -1 for the "end of line" case (supposedly a value that no
947        character can have). Though this value is never used (the check for end of
948        line is "zero bytes in current character"), it caused compiler complaints.
949        I've changed it to 0xffffffff.
950    
951     8. In pcre_version.c, the version string was being built by a sequence of
952        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
953        string (as it is for production releases) called a macro with an empty
954        argument. The C standard says the result of this is undefined. The gcc
955        compiler treats it as an empty string (which was what was wanted) but it is
956        reported that Visual C gives an error. The source has been hacked around to
957        avoid this problem.
958    
959     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
960        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
961        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
962        of them did).
963    
964    10. Originally, pcretest opened its input and output without "b"; then I was
965        told that "b" was needed in some environments, so it was added for release
966        5.0 to both the input and output. (It makes no difference on Unix-like
967        systems.) Later I was told that it is wrong for the input on Windows. I've
968        now abstracted the modes into two macros, to make it easier to fiddle with
969        them, and removed "b" from the input mode under Windows.
970    
971    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
972    
973    12. Added -help and --help to pcretest as an official way of being reminded
974        of the options.
975    
976    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
977        and pcrecpp.cc because they annoy compilers at high warning levels.
978    
979    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
980    
981    15. Fixed an occurrence of == in configure.ac that should have been = (shell
982        scripts are not C programs :-) and which was not noticed because it works
983        on Linux.
984    
985    16. pcretest is supposed to handle any length of pattern and data line (as one
986        line or as a continued sequence of lines) by extending its input buffer if
987        necessary. This feature was broken for very long pattern lines, leading to
988        a string of junk being passed to pcre_compile() if the pattern was longer
989        than about 50K.
990    
991    17. I have done a major re-factoring of the way pcre_compile() computes the
992        amount of memory needed for a compiled pattern. Previously, there was code
993        that made a preliminary scan of the pattern in order to do this. That was
994        OK when PCRE was new, but as the facilities have expanded, it has become
995        harder and harder to keep it in step with the real compile phase, and there
996        have been a number of bugs (see for example, 4 above). I have now found a
997        cunning way of running the real compile function in a "fake" mode that
998        enables it to compute how much memory it would need, while actually only
999        ever using a few hundred bytes of working memory and without too many
1000        tests of the mode. This should make future maintenance and development
1001        easier. A side effect of this work is that the limit of 200 on the nesting
1002        depth of parentheses has been removed (though this was never a serious
1003        limitation, I suspect). However, there is a downside: pcre_compile() now
1004        runs more slowly than before (30% or more, depending on the pattern). I
1005        hope this isn't a big issue. There is no effect on runtime performance.
1006    
1007    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1008        newline (only possible for the last line of a file) and it was a
1009        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1010    
1011    19. Added additional timing features to pcretest. (1) The -tm option now times
1012        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1013        separate command line item, by a number that specifies the number of
1014        repeats to use when timing. The default is 50000; this gives better
1015        precision, but takes uncomfortably long for very large patterns.
1016    
1017    20. Extended pcre_study() to be more clever in cases where a branch of a
1018        subpattern has no definite first character. For example, (a*|b*)[cd] would
1019        previously give no result from pcre_study(). Now it recognizes that the
1020        first character must be a, b, c, or d.
1021    
1022    21. There was an incorrect error "recursive call could loop indefinitely" if
1023        a subpattern (or the entire pattern) that was being tested for matching an
1024        empty string contained only one non-empty item after a nested subpattern.
1025        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1026        incorrectly, because the \d was being skipped in the check.
1027    
1028    22. The pcretest program now has a new pattern option /B and a command line
1029        option -b, which is equivalent to adding /B to every pattern. This causes
1030        it to show the compiled bytecode, without the additional information that
1031        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1032        is the same as /B/I).
1033    
1034    23. A new optimization is now able automatically to treat some sequences such
1035        as a*b as a*+b. More specifically, if something simple (such as a character
1036        or a simple class like \d) has an unlimited quantifier, and is followed by
1037        something that cannot possibly match the quantified thing, the quantifier
1038        is automatically "possessified".
1039    
1040    24. A recursive reference to a subpattern whose number was greater than 39
1041        went wrong under certain circumstances in UTF-8 mode. This bug could also
1042        have affected the operation of pcre_study().
1043    
1044    25. Realized that a little bit of performance could be had by replacing
1045        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1046    
1047    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1048    
1049    27. Possessive quantifiers such as a++ were previously implemented by turning
1050        them into atomic groups such as ($>a+). Now they have their own opcodes,
1051        which improves performance. This includes the automatically created ones
1052        from 23 above.
1053    
1054    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1055        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1056        the first matched character to be a colon. This applied both to named and
1057        numbered groups.
1058    
1059    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1060    
1061    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1062        building PCRE on Windows easier, so I have included it in the distribution.
1063    
1064    31. There is now a check in pcretest against a ridiculously large number being
1065        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1066        loop, the loop is abandoned.
1067    
1068    32. Forward references to subpatterns in conditions such as (?(2)...) where
1069        subpattern 2 is defined later cause pcre_compile() to search forwards in
1070        the pattern for the relevant set of parentheses. This search went wrong
1071        when there were unescaped parentheses in a character class, parentheses
1072        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1073    
1074    33. "Subroutine" calls and backreferences were previously restricted to
1075        referencing subpatterns earlier in the regex. This restriction has now
1076        been removed.
1077    
1078    34. Added a number of extra features that are going to be in Perl 5.10. On the
1079        whole, these are just syntactic alternatives for features that PCRE had
1080        previously implemented using the Python syntax or my own invention. The
1081        other formats are all retained for compatibility.
1082    
1083        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1084            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1085            also .NET compatible.
1086    
1087        (b) A recursion or subroutine call to a named group can now be defined as
1088            (?&name) as well as (?P>name).
1089    
1090        (c) A backreference to a named group can now be defined as \k<name> or
1091            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1092            5.10, are also .NET compatible.
1093    
1094        (d) A conditional reference to a named group can now use the syntax
1095            (?(<name>) or (?('name') as well as (?(name).
1096    
1097        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1098            groups (named and numbered) that are never evaluated inline, but can be
1099            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1100            is always false. There may be only one alternative in such a group.
1101    
1102        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1103            as the simple (?(R). The condition is true only if the most recent
1104            recursion is that of the given number or name. It does not search out
1105            through the entire recursion stack.
1106    
1107        (g) The escape \gN or \g{N} has been added, where N is a positive or
1108            negative number, specifying an absolute or relative reference.
1109    
1110    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1111        some "unreachable code" warnings.
1112    
1113    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1114        things, this adds five new scripts.
1115    
1116    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1117        There were also incompatibilities regarding the handling of \Q..\E inside
1118        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1119        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1120    
1121    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1122        matches an empty string, and forcibly breaks the loop. There were bugs in
1123        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1124        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1125        separate and independent bugs (that affected different cases) have been
1126        fixed.
1127    
1128    39. Refactored the code to abolish the use of different opcodes for small
1129        capturing bracket numbers. This is a tidy that I avoided doing when I
1130        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1131        The new approach is not only tidier, it makes it possible to reduce the
1132        memory needed to fix the previous bug (38).
1133    
1134    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1135        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1136        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1137        mode.
1138    
1139    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1140        report.
1141    
1142    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1143        copy construction and assignment in the C++ wrapper.
1144    
1145    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1146        couple of bugs that could have given wrong results in the "--newline=crlf"
1147        case.
1148    
1149    44. Added a number of casts and did some reorganization of signed/unsigned int
1150        variables following suggestions from Dair Grant. Also renamed the variable
1151        "this" as "item" because it is a C++ keyword.
1152    
1153    45. Arranged for dftables to add
1154    
1155          #include "pcre_internal.h"
1156    
1157        to pcre_chartables.c because without it, gcc 4.x may remove the array
1158        definition from the final binary if PCRE is built into a static library and
1159        dead code stripping is activated.
1160    
1161    46. For an unanchored pattern, if a match attempt fails at the start of a
1162        newline sequence, and the newline setting is CRLF or ANY, and the next two
1163        characters are CRLF, advance by two characters instead of one.
1164    
1165    
1166    Version 6.7 04-Jul-06
1167    ---------------------
1168    
1169     1. In order to handle tests when input lines are enormously long, pcretest has
1170        been re-factored so that it automatically extends its buffers when
1171        necessary. The code is crude, but this _is_ just a test program. The
1172        default size has been increased from 32K to 50K.
1173    
1174     2. The code in pcre_study() was using the value of the re argument before
1175        testing it for NULL. (Of course, in any sensible call of the function, it
1176        won't be NULL.)
1177    
1178     3. The memmove() emulation function in pcre_internal.h, which is used on
1179        systems that lack both memmove() and bcopy() - that is, hardly ever -
1180        was missing a "static" storage class specifier.
1181    
1182     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1183        containing an extended class (one that cannot be represented by a bitmap
1184        because it contains high-valued characters or Unicode property items, e.g.
1185        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1186        pattern, but PCRE should not loop if you do not (it no longer does).
1187        [Detail: two cases were found: (a) a repeated subpattern containing an
1188        extended class; (b) a recursive reference to a subpattern that followed a
1189        previous extended class. It wasn't skipping over the extended class
1190        correctly when UTF-8 mode was not set.]
1191    
1192     5. A negated single-character class was not being recognized as fixed-length
1193        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1194        compile error "lookbehind assertion is not fixed length".
1195    
1196     6. The RunPerlTest auxiliary script was showing an unexpected difference
1197        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1198        write a Perl script that can interpret lines of an input file either as
1199        byte characters or as UTF-8, which is what "perltest" was being required to
1200        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1201        can't do is switch easily at run time between having the "use utf8;" pragma
1202        or not. In the end, I fudged it by using the RunPerlTest script to insert
1203        "use utf8;" explicitly for the UTF-8 tests.
1204    
1205     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1206        the end of the subject string, contrary to the documentation and to what
1207        Perl does. This was true of both matching functions. Now it matches only at
1208        the start of the subject and immediately after *internal* newlines.
1209    
1210     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1211        a pointer to an int instead of a pointer to an unsigned long int. This
1212        caused problems on 64-bit systems.
1213    
1214     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1215        instance of the 'standard' template library not being so standard".
1216    
1217    10. There was no check on the number of named subpatterns nor the maximum
1218        length of a subpattern name. The product of these values is used to compute
1219        the size of the memory block for a compiled pattern. By supplying a very
1220        long subpattern name and a large number of named subpatterns, the size
1221        computation could be caused to overflow. This is now prevented by limiting
1222        the length of names to 32 characters, and the number of named subpatterns
1223        to 10,000.
1224    
1225    11. Subpatterns that are repeated with specific counts have to be replicated in
1226        the compiled pattern. The size of memory for this was computed from the
1227        length of the subpattern and the repeat count. The latter is limited to
1228        65535, but there was no limit on the former, meaning that integer overflow
1229        could in principle occur. The compiled length of a repeated subpattern is
1230        now limited to 30,000 bytes in order to prevent this.
1231    
1232    12. Added the optional facility to have named substrings with the same name.
1233    
1234    13. Added the ability to use a named substring as a condition, using the
1235        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1236        are numbers (not recommended). Forward references are permitted.
1237    
1238    14. Added forward references in named backreferences (if you see what I mean).
1239    
1240    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1241        pattern could run off the end of the subject. For example, the pattern
1242        "(?s)(.{1,5})"8 did this with the subject "ab".
1243    
1244    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1245        PCRE_CASELESS was set when matching characters that were quantified with ?
1246        or *.
1247    
1248    17. A character class other than a single negated character that had a minimum
1249        but no maximum quantifier - for example [ab]{6,} - was not handled
1250        correctly by pce_dfa_exec(). It would match only one character.
1251    
1252    18. A valid (though odd) pattern that looked like a POSIX character
1253        class but used an invalid character after [ (for example [[,abc,]]) caused
1254        pcre_compile() to give the error "Failed: internal error: code overflow" or
1255        in some cases to crash with a glibc free() error. This could even happen if
1256        the pattern terminated after [[ but there just happened to be a sequence of
1257        letters, a binary zero, and a closing ] in the memory that followed.
1258    
1259    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1260        over the years. Originally (before any Unicode support), just the bottom 8
1261        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1262        output from "man perlunicode" includes this:
1263    
1264          The regular expression compiler produces polymorphic opcodes.  That
1265          is, the pattern adapts to the data and automatically switches to
1266          the Unicode character scheme when presented with Unicode data--or
1267          instead uses a traditional byte scheme when presented with byte
1268          data.
1269    
1270        Sadly, a wide octal escape does not cause a switch, and in a string with
1271        no other multibyte characters, these octal escapes are treated as before.
1272        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1273        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1274        Unicode string.
1275    
1276        I have not perpetrated such confusion in PCRE. Up till now, it took just
1277        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1278        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1279        translate to the appropriate multibyte character.
1280    
1281    29. Applied some refactoring to reduce the number of warnings from Microsoft
1282        and Borland compilers. This has included removing the fudge introduced
1283        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1284        a warning about an unused variable.
1285    
1286    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1287        characters since release 4.0, because Perl (from release 5.004) does not.
1288        [Or at least, is documented not to: some releases seem to be in conflict
1289        with the documentation.] However, when a pattern was studied with
1290        pcre_study() and all its branches started with \s, PCRE still included VT
1291        as a possible starting character. Of course, this did no harm; it just
1292        caused an unnecessary match attempt.
1293    
1294    22. Removed a now-redundant internal flag bit that recorded the fact that case
1295        dependency changed within the pattern. This was once needed for "required
1296        byte" processing, but is no longer used. This recovers a now-scarce options
1297        bit. Also moved the least significant internal flag bit to the most-
1298        significant bit of the word, which was not previously used (hangover from
1299        the days when it was an int rather than a uint) to free up another bit for
1300        the future.
1301    
1302    23. Added support for CRLF line endings as well as CR and LF. As well as the
1303        default being selectable at build time, it can now be changed at runtime
1304        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1305        specify that it is scanning data with non-default line endings.
1306    
1307    24. Changed the definition of CXXLINK to make it agree with the definition of
1308        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1309    
1310    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1311        recursions. This makes a big different to stack usage for some patterns.
1312    
1313    26. If a subpattern containing a named recursion or subroutine reference such
1314        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1315        the space required for the compiled pattern went wrong and gave too small a
1316        value. Depending on the environment, this could lead to "Failed: internal
1317        error: code overflow at offset 49" or "glibc detected double free or
1318        corruption" errors.
1319    
1320    27. Applied patches from Google (a) to support the new newline modes and (b) to
1321        advance over multibyte UTF-8 characters in GlobalReplace.
1322    
1323    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1324        difference for some implementation of PCRE in some Windows version.
1325    
1326    29. Added some extra testing facilities to pcretest:
1327    
1328        \q<number>   in a data line sets the "match limit" value
1329        \Q<number>   in a data line sets the "match recursion limt" value
1330        -S <number>  sets the stack size, where <number> is in megabytes
1331    
1332        The -S option isn't available for Windows.
1333    
1334    
1335    Version 6.6 06-Feb-06
1336    ---------------------
1337    
1338     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1339        in pcreposix.h. I have copied the definition from pcre.h.
1340    
1341     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1342        because pcre.h is no longer a built file.
1343    
1344     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1345        not normally included in the compiled code.
1346    
1347    
1348    Version 6.5 01-Feb-06
1349    ---------------------
1350    
1351     1. When using the partial match feature with pcre_dfa_exec(), it was not
1352        anchoring the second and subsequent partial matches at the new starting
1353        point. This could lead to incorrect results. For example, with the pattern
1354        /1234/, partially matching against "123" and then "a4" gave a match.
1355    
1356     2. Changes to pcregrep:
1357    
1358        (a) All non-match returns from pcre_exec() were being treated as failures
1359            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1360            error message is output. Some extra information is given for the
1361            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1362            probably the only errors that are likely to be caused by users (by
1363            specifying a regex that has nested indefinite repeats, for instance).
1364            If there are more than 20 of these errors, pcregrep is abandoned.
1365    
1366        (b) A binary zero was treated as data while matching, but terminated the
1367            output line if it was written out. This has been fixed: binary zeroes
1368            are now no different to any other data bytes.
1369    
1370        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1371            used to set a locale for matching. The --locale=xxxx long option has
1372            been added (no short equivalent) to specify a locale explicitly on the
1373            pcregrep command, overriding the environment variables.
1374    
1375        (d) When -B was used with -n, some line numbers in the output were one less
1376            than they should have been.
1377    
1378        (e) Added the -o (--only-matching) option.
1379    
1380        (f) If -A or -C was used with -c (count only), some lines of context were
1381            accidentally printed for the final match.
1382    
1383        (g) Added the -H (--with-filename) option.
1384    
1385        (h) The combination of options -rh failed to suppress file names for files
1386            that were found from directory arguments.
1387    
1388        (i) Added the -D (--devices) and -d (--directories) options.
1389    
1390        (j) Added the -F (--fixed-strings) option.
1391    
1392        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1393    
1394        (l) Added the --colo(u)r option.
1395    
1396        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1397            is not present by default.
1398    
1399     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1400        items such as (?R) or (?1), when the recursion could match a number of
1401        alternatives. If it matched one of the alternatives, but subsequently,
1402        outside the recursion, there was a failure, the code tried to back up into
1403        the recursion. However, because of the way PCRE is implemented, this is not
1404        possible, and the result was an incorrect result from the match.
1405    
1406        In order to prevent this happening, the specification of recursion has
1407        been changed so that all such subpatterns are automatically treated as
1408        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1409    
1410     4. I had overlooked the fact that, in some locales, there are characters for
1411        which isalpha() is true but neither isupper() nor islower() are true. In
1412        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1413        and ordfeminine) are like this. This affected the treatment of \w and \W
1414        when they appeared in character classes, but not when they appeared outside
1415        a character class. The bit map for "word" characters is now created
1416        separately from the results of isalnum() instead of just taking it from the
1417        upper, lower, and digit maps. (Plus the underscore character, of course.)
1418    
1419     5. The above bug also affected the handling of POSIX character classes such as
1420        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1421        permanent tables. Instead, the bit maps for such a class were previously
1422        created as the appropriate unions of the upper, lower, and digit bitmaps.
1423        Now they are created by subtraction from the [[:word:]] class, which has
1424        its own bitmap.
1425    
1426     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1427        It is created by subtracting the vertical space characters (\x09, \x0a,
1428        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1429        subtraction was done in the overall bitmap for a character class, meaning
1430        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1431        be recognized. This bug has been fixed.
1432    
1433     7. Patches from the folks at Google:
1434    
1435          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1436          real life, but is still worth protecting against".
1437    
1438          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1439          regular expressions".
1440    
1441          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1442          have it.
1443    
1444          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1445          "configure" and the latter not, in order to fix a problem somebody had
1446          with compiling the Arg class on HP-UX.
1447    
1448          (e) Improve the error-handling of the C++ wrapper a little bit.
1449    
1450          (f) New tests for checking recursion limiting.
1451    
1452     8. The pcre_memmove() function, which is used only if the environment does not
1453        have a standard memmove() function (and is therefore rarely compiled),
1454        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1455        returning a result (though PCRE never actually uses the result).
1456    
1457     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1458        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1459        returned instead of calling malloc() with an overflowing number that would
1460        most likely cause subsequent chaos.
1461    
1462    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1463    
1464    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1465        with this option is matched, the nmatch and pmatch options of regexec() are
1466        ignored.
1467    
1468    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1469        provided in case anyone wants to the the POSIX interface with UTF-8
1470        strings.
1471    
1472    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1473        C++ linking (needed for some HP-UX environments).
1474    
1475    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1476        (unused parameter) and in the pcre_printint() function (omitted "default"
1477        switch label when the default is to do nothing).
1478    
1479    15. Added some code to make it possible, when PCRE is compiled as a C++
1480        library, to replace subject pointers for pcre_exec() with a smart pointer
1481        class, thus making it possible to process discontinuous strings.
1482    
1483    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1484        much the same function. They were added by different people who were trying
1485        to make PCRE easy to compile on non-Unix systems. It has been suggested
1486        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1487        for compiling on Windows systems. I have therefore replaced it with
1488        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1489        defaults to "extern" for C or "extern C" for C++, which works fine on
1490        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1491        SCOPE with something explicit in config.h. In addition:
1492    
1493        (a) pcreposix.h still had just "extern" instead of either of these macros;
1494            I have replaced it with PCRE_DATA_SCOPE.
1495    
1496        (b) Functions such as _pcre_xclass(), which are internal to the library,
1497            but external in the C sense, all had PCRE_EXPORT in their definitions.
1498            This is apparently wrong for the Windows case, so I have removed it.
1499            (It makes no difference on Unix-like systems.)
1500    
1501    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1502        of recursive calls to match(). This is different to MATCH_LIMIT because
1503        that limits the total number of calls to match(), not all of which increase
1504        the depth of recursion. Limiting the recursion depth limits the amount of
1505        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1506        when PCRE is compiled, and changed at run time. A patch from Google adds
1507        this functionality to the C++ interface.
1508    
1509    18. Changes to the handling of Unicode character properties:
1510    
1511        (a) Updated the table to Unicode 4.1.0.
1512    
1513        (b) Recognize characters that are not in the table as "Cn" (undefined).
1514    
1515        (c) I revised the way the table is implemented to a much improved format
1516            which includes recognition of ranges. It now supports the ranges that
1517            are defined in UnicodeData.txt, and it also amalgamates other
1518            characters into ranges. This has reduced the number of entries in the
1519            table from around 16,000 to around 3,000, thus reducing its size
1520            considerably. I realized I did not need to use a tree structure after
1521            all - a binary chop search is just as efficient. Having reduced the
1522            number of entries, I extended their size from 6 bytes to 8 bytes to
1523            allow for more data.
1524    
1525        (d) Added support for Unicode script names via properties such as \p{Han}.
1526    
1527    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1528        matching that character.
1529    
1530    20. When matching a repeated Unicode property with a minimum greater than zero,
1531        (for example \pL{2,}), PCRE could look past the end of the subject if it
1532        reached it while seeking the minimum number of characters. This could
1533        happen only if some of the characters were more than one byte long, because
1534        there is a check for at least the minimum number of bytes.
1535    
1536    21. Refactored the implementation of \p and \P so as to be more general, to
1537        allow for more different types of property in future. This has changed the
1538        compiled form incompatibly. Anybody with saved compiled patterns that use
1539        \p or \P will have to recompile them.
1540    
1541    22. Added "Any" and "L&" to the supported property types.
1542    
1543    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1544        but give a compile time error if the value is greater than 0xff.
1545    
1546    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1547        accidentally not being installed or uninstalled.
1548    
1549    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1550        made were to insert the current release number. This seemed silly, because
1551        it made things harder for people building PCRE on systems that don't run
1552        "configure". I have turned pcre.h into a distributed file, no longer built
1553        by "configure", with the version identification directly included. There is
1554        no longer a pcre.h.in file.
1555    
1556        However, this change necessitated a change to the pcre-config script as
1557        well. It is built from pcre-config.in, and one of the substitutions was the
1558        release number. I have updated configure.ac so that ./configure now finds
1559        the release number by grepping pcre.h.
1560    
1561    26. Added the ability to run the tests under valgrind.
1562    
1563    
1564    Version 6.4 05-Sep-05
1565    ---------------------
1566    
1567     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1568        "--" to be printed when multiple files were scanned, even when none of the
1569        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1570        consider it to be a bug, and have restored the previous behaviour.
1571    
1572     2. A couple of code tidies to get rid of compiler warnings.
1573    
1574     3. The pcretest program used to cheat by referring to symbols in the library
1575        whose names begin with _pcre_. These are internal symbols that are not
1576        really supposed to be visible externally, and in some environments it is
1577        possible to suppress them. The cheating is now confined to including
1578        certain files from the library's source, which is a bit cleaner.
1579    
1580     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1581        file's purpose clearer.
1582    
1583     5. Reorganized pcre_ucp_findchar().
1584    
1585    
1586    Version 6.3 15-Aug-05
1587    ---------------------
1588    
1589     1. The file libpcre.pc.in did not have general read permission in the tarball.
1590    
1591     2. There were some problems when building without C++ support:
1592    
1593        (a) If C++ support was not built, "make install" and "make test" still
1594            tried to test it.
1595    
1596        (b) There were problems when the value of CXX was explicitly set. Some
1597            changes have been made to try to fix these, and ...
1598    
1599        (c) --disable-cpp can now be used to explicitly disable C++ support.
1600    
1601        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1602            backslash in a target when C++ was disabled. This confuses some
1603            versions of "make", apparently. Using an intermediate variable solves
1604            this. (Same for CPP_LOBJ.)
1605    
1606     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1607        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1608        necessary on certain architectures.
1609    
1610     4. Added a setting of -export-symbols-regex to the link command to remove
1611        those symbols that are exported in the C sense, but actually are local
1612        within the library, and not documented. Their names all begin with
1613        "_pcre_". This is not a perfect job, because (a) we have to except some
1614        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1615        available (and never for static libraries). I have made a note to try to
1616        find a way round (a) in the future.
1617    
1618    
1619    Version 6.2 01-Aug-05
1620    ---------------------
1621    
1622     1. There was no test for integer overflow of quantifier values. A construction
1623        such as {1111111111111111} would give undefined results. What is worse, if
1624        a minimum quantifier for a parenthesized subpattern overflowed and became
1625        negative, the calculation of the memory size went wrong. This could have
1626        led to memory overwriting.
1627    
1628     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1629    
1630     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1631        operating environments where this matters.
1632    
1633     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1634        PCRE options from within the C++ wrapper.
1635    
1636     5. Named capturing subpatterns were not being correctly counted when a pattern
1637        was compiled. This caused two problems: (a) If there were more than 100
1638        such subpatterns, the calculation of the memory needed for the whole
1639        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1640        back references of the form \12, where the number was greater than 9, were
1641        not recognized as back references, even though there were sufficient
1642        previous subpatterns.
1643    
1644     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1645        versions of gcc, e.g. 2.95.4.
1646    
1647    
1648    Version 6.1 21-Jun-05
1649    ---------------------
1650    
1651     1. There was one reference to the variable "posix" in pcretest.c that was not
1652        surrounded by "#if !defined NOPOSIX".
1653    
1654     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1655        the cross-check on the old pcre_info() function, for the benefit of the
1656        cut-down version of PCRE that is currently imported into Exim.
1657    
1658     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1659        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1660        patterns that start (?i) but I don't think that matters. The use of (?i) is
1661        just an example; this all applies to the other options as well.
1662    
1663     4. Since libtool seems to echo the compile commands it is issuing, the output
1664        from "make" can be reduced a bit by putting "@" in front of each libtool
1665        compile command.
1666    
1667     5. Patch from the folks at Google for configure.in to be a bit more thorough
1668        in checking for a suitable C++ installation before trying to compile the
1669        C++ stuff. This should fix a reported problem when a compiler was present,
1670        but no suitable headers.
1671    
1672     6. The man pages all had just "PCRE" as their title. I have changed them to
1673        be the relevant file name. I have also arranged that these names are
1674        retained in the file doc/pcre.txt, which is a concatenation in text format
1675        of all the man pages except the little individual ones for each function.
1676    
1677     7. The NON-UNIX-USE file had not been updated for the different set of source
1678        files that come with release 6. I also added a few comments about the C++
1679        wrapper.
1680    
1681    
1682    Version 6.0 07-Jun-05
1683    ---------------------
1684    
1685     1. Some minor internal re-organization to help with my DFA experiments.
1686    
1687     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1688        didn't matter for the library itself when fully configured, but did matter
1689        when compiling without UCP support, or within Exim, where the ucp files are
1690        not imported.
1691    
1692     3. Refactoring of the library code to split up the various functions into
1693        different source modules. The addition of the new DFA matching code (see
1694        below) to a single monolithic source would have made it really too
1695        unwieldy, quite apart from causing all the code to be include in a
1696        statically linked application, when only some functions are used. This is
1697        relevant even without the DFA addition now that patterns can be compiled in
1698        one application and matched in another.
1699    
1700        The downside of splitting up is that there have to be some external
1701        functions and data tables that are used internally in different modules of
1702        the library but which are not part of the API. These have all had their
1703        names changed to start with "_pcre_" so that they are unlikely to clash
1704        with other external names.
1705    
1706     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1707        a different (DFA) algorithm. Although it is slower than the original
1708        function, it does have some advantages for certain types of matching
1709        problem.
1710    
1711     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1712        including restarting after a partial match.
1713    
1714     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1715        defined when compiling for Windows was sent to me. I have put it into the
1716        code, though I have no means of testing or verifying it.
1717    
1718     7. Added the pcre_refcount() auxiliary function.
1719    
1720     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1721        match before or at the first newline in the subject string. In pcretest,
1722        the /f option on a pattern can be used to set this.
1723    
1724     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1725        would behave wrongly. This has been present in PCRE since release 4.0.
1726    
1727    10. A number of changes to the pcregrep command:
1728    
1729        (a) Refactored how -x works; insert ^(...)$ instead of setting
1730            PCRE_ANCHORED and checking the length, in preparation for adding
1731            something similar for -w.
1732    
1733        (b) Added the -w (match as a word) option.
1734    
1735        (c) Refactored the way lines are read and buffered so as to have more
1736            than one at a time available.
1737    
1738        (d) Implemented a pcregrep test script.
1739    
1740        (e) Added the -M (multiline match) option. This allows patterns to match
1741            over several lines of the subject. The buffering ensures that at least
1742            8K, or the rest of the document (whichever is the shorter) is available
1743            for matching (and similarly the previous 8K for lookbehind assertions).
1744    
1745        (f) Changed the --help output so that it now says
1746    
1747              -w, --word-regex(p)
1748    
1749            instead of two lines, one with "regex" and the other with "regexp"
1750            because that confused at least one person since the short forms are the
1751            same. (This required a bit of code, as the output is generated
1752            automatically from a table. It wasn't just a text change.)
1753    
1754        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1755            option but starts with a hyphen. Could be a pattern or a path name
1756            starting with a hyphen, for instance.
1757    
1758        (h) "-" can be given as a file name to represent stdin.
1759    
1760        (i) When file names are being printed, "(standard input)" is used for
1761            the standard input, for compatibility with GNU grep. Previously
1762            "<stdin>" was used.
1763    
1764        (j) The option --label=xxx can be used to supply a name to be used for
1765            stdin when file names are being printed. There is no short form.
1766    
1767        (k) Re-factored the options decoding logic because we are going to add
1768            two more options that take data. Such options can now be given in four
1769            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1770    
1771        (l) Added the -A, -B, and -C options for requesting that lines of context
1772            around matches be printed.
1773    
1774        (m) Added the -L option to print the names of files that do not contain
1775            any matching lines, that is, the complement of -l.
1776    
1777        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1778            continue to scan other files.
1779    
1780        (o) The -s option was incorrectly implemented. For compatibility with other
1781            greps, it now suppresses the error message for a non-existent or non-
1782            accessible file (but not the return code). There is a new option called
1783            -q that suppresses the output of matching lines, which was what -s was
1784            previously doing.
1785    
1786        (p) Added --include and --exclude options to specify files for inclusion
1787            and exclusion when recursing.
1788    
1789    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1790        Hopefully, it now does.
1791    
1792    12. Missing cast in pcre_study().
1793    
1794    13. Added an "uninstall" target to the makefile.
1795    
1796    14. Replaced "extern" in the function prototypes in Makefile.in with
1797        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1798        world, but is set differently for Windows.
1799    
1800    15. Added a second compiling function called pcre_compile2(). The only
1801        difference is that it has an extra argument, which is a pointer to an
1802        integer error code. When there is a compile-time failure, this is set
1803        non-zero, in addition to the error test pointer being set to point to an
1804        error message. The new argument may be NULL if no error number is required
1805        (but then you may as well call pcre_compile(), which is now just a
1806        wrapper). This facility is provided because some applications need a
1807        numeric error indication, but it has also enabled me to tidy up the way
1808        compile-time errors are handled in the POSIX wrapper.
1809    
1810    16. Added VPATH=.libs to the makefile; this should help when building with one
1811        prefix path and installing with another. (Or so I'm told by someone who
1812        knows more about this stuff than I do.)
1813    
1814    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1815        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1816        match everything, including newlines. This is not POSIX-compatible, but
1817        somebody wanted the feature. From pcretest it can be activated by using
1818        both the P and the s flags.
1819    
1820    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1821    
1822    19. libpcre.pc was being incorrectly installed as executable.
1823    
1824    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1825        it now also looks for '\r' so that it will work unmodified on Windows.
1826    
1827    21. Added Google's contributed C++ wrapper to the distribution.
1828    
1829    22. Added some untidy missing memory free() calls in pcretest, to keep
1830        Electric Fence happy when testing.
1831    
1832    
1833    
1834    Version 5.0 13-Sep-04
1835    ---------------------
1836    
1837     1. Internal change: literal characters are no longer packed up into items
1838        containing multiple characters in a single byte-string. Each character
1839        is now matched using a separate opcode. However, there may be more than one
1840        byte in the character in UTF-8 mode.
1841    
1842     2. The pcre_callout_block structure has two new fields: pattern_position and
1843        next_item_length. These contain the offset in the pattern to the next match
1844        item, and its length, respectively.
1845    
1846     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1847        insertion of callouts before each pattern item. Added the /C option to
1848        pcretest to make use of this.
1849    
1850     4. On the advice of a Windows user, the lines
1851    
1852          #if defined(_WIN32) || defined(WIN32)
1853          _setmode( _fileno( stdout ), 0x8000 );
1854          #endif  /* defined(_WIN32) || defined(WIN32) */
1855    
1856        have been added to the source of pcretest. This apparently does useful
1857        magic in relation to line terminators.
1858    
1859     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1860        for the benefit of those environments where the "b" makes a difference.
1861    
1862     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1863        to know about it. I have put a hack into configure.in that adds in code
1864        to set GCC=yes if CC=icc. This seems to end up at a point in the
1865        generated configure script that is early enough to affect the setting of
1866        compiler options, which is what is needed, but I have no means of testing
1867        whether it really works. (The user who reported this had patched the
1868        generated configure script, which of course I cannot do.)
1869    
1870        LATER: After change 22 below (new libtool files), the configure script
1871        seems to know about icc (and also ecc). Therefore, I have commented out
1872        this hack in configure.in.
1873    
1874     7. Added support for pkg-config (2 patches were sent in).
1875    
1876     8. Negated POSIX character classes that used a combination of internal tables
1877        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1878        [[:^ascii]]. Typically, they would match almost any characters. The other
1879        POSIX classes were not broken in this way.
1880    
1881     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1882        to find the match, as PCRE was deluded into thinking that the match had to
1883        start at the start point or following a newline. The same bug applied to
1884        patterns with negative forward assertions or any backward assertions
1885        preceding ".*" at the start, unless the pattern required a fixed first
1886        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1887    
1888    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1889        starting at the last subject character, bytes beyond the end of the subject
1890        string were read.
1891    
1892    11. Renamed the variable "class" as "classbits" to make life easier for C++
1893        users. (Previously there was a macro definition, but it apparently wasn't
1894        enough.)
1895    
1896    12. Added the new field "tables" to the extra data so that tables can be passed
1897        in at exec time, or the internal tables can be re-selected. This allows
1898        a compiled regex to be saved and re-used at a later time by a different
1899        program that might have everything at different addresses.
1900    
1901    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1902        -R library as well as a -L library.
1903    
1904    14. The debugging options of pcretest (-d on the command line or D on a
1905        pattern) showed incorrect output for anything following an extended class
1906        that contained multibyte characters and which was followed by a quantifier.
1907    
1908    15. Added optional support for general category Unicode character properties
1909        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1910        support. It adds about 90K to the size of the library. The meanings of the
1911        inbuilt class escapes such as \d and \s have NOT been changed.
1912    
1913    16. Updated pcredemo.c to include calls to free() to release the memory for the
1914        compiled pattern.
1915    
1916    17. The generated file chartables.c was being created in the source directory
1917        instead of in the building directory. This caused the build to fail if the
1918        source directory was different from the building directory, and was
1919        read-only.
1920    
1921    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1922        file. No doubt somebody will tell me if they don't make sense... Also added
1923        Dan Mooney's comments about building on OpenVMS.
1924    
1925    19. Added support for partial matching via the PCRE_PARTIAL option for
1926        pcre_exec() and the \P data escape in pcretest.
1927    
1928    20. Extended pcretest with 3 new pattern features:
1929    
1930        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1931              write the compiled pattern to the file whose name is "rest-of-line".
1932              This is a straight binary dump of the data, with the saved pointer to
1933              the character tables forced to be NULL. The study data, if any, is
1934              written too. After writing, pcretest reads a new pattern.
1935    
1936        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1937              compiled pattern from the given file. There must not be any
1938              occurrences of "<" in the file name (pretty unlikely); if there are,
1939              pcretest will instead treat the initial "<" as a pattern delimiter.
1940              After reading in the pattern, pcretest goes on to read data lines as
1941              usual.
1942    
1943        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1944              and 16-bit fields in a compiled pattern, to simulate a pattern that
1945              was compiled on a host of opposite endianness.
1946    
1947    21. The pcre-exec() function can now cope with patterns that were compiled on
1948        hosts of opposite endianness, with this restriction:
1949    
1950          As for any compiled expression that is saved and used later, the tables
1951          pointer field cannot be preserved; the extra_data field in the arguments
1952          to pcre_exec() should be used to pass in a tables address if a value
1953          other than the default internal tables were used at compile time.
1954    
1955    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1956        now diagnosed as an error. Previously, most of the time, a negative number
1957        would have been treated as zero, but if in addition "ovector" was passed as
1958        NULL, a crash could occur.
1959    
1960    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1961        new versions from the libtool 1.5 distribution (the last one is a copy of
1962        a file called libtool.m4). This seems to have fixed the need to patch
1963        "configure" to support Darwin 1.3 (which I used to do). However, I still
1964        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1965        workstation).
1966    
1967    24. Changed the PCRE licence to be the more standard "BSD" licence.
1968    
1969    
1970    Version 4.5 01-Dec-03
1971    ---------------------
1972    
1973     1. There has been some re-arrangement of the code for the match() function so
1974        that it can be compiled in a version that does not call itself recursively.
1975        Instead, it keeps those local variables that need separate instances for
1976        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1977        needs to "recurse". Keeping track of where control must go is done by means
1978        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1979        hide most of the details from the main code, and operates only if
1980        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1981        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1982        operating.
1983    
1984        To make it easier for callers to provide specially tailored get/free
1985        functions for this usage, two new functions, pcre_stack_malloc, and
1986        pcre_stack_free, are used. They are always called in strict stacking order,
1987        and the size of block requested is always the same.
1988    
1989        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1990        PCRE has been compiled to use the stack or the heap for recursion. The
1991        -C option of pcretest uses this to show which version is compiled.
1992    
1993        A new data escape \S, is added to pcretest; it causes the amounts of store
1994        obtained and freed by both kinds of malloc/free at match time to be added
1995        to the output.
1996    
1997     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1998        what's available on my current Linux desktop machine.
1999    
2000     3. When matching a UTF-8 string, the test for a valid string at the start has
2001        been extended. If start_offset is not zero, PCRE now checks that it points
2002        to a byte that is the start of a UTF-8 character. If not, it returns
2003        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2004        this is necessary because there may be backward assertions in the pattern.
2005        When matching the same subject several times, it may save resources to use
2006        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2007    
2008     4. The code for checking the validity of UTF-8 strings has been tightened so
2009        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2010        containing "overlong sequences".
2011    
2012     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2013        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2014        should have been "&", but it just so happened that all the cases this let
2015        through by mistake were picked up later in the function.
2016    
2017     6. I had used a variable called "isblank" - this is a C99 function, causing
2018        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2019    
2020     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2021        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2022        so the version is shown; (c) stop "make test" echoing "./RunTest".
2023    
2024     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2025    
2026     9. The prototype for memmove() for systems that don't have it was using
2027        size_t, but the inclusion of the header that defines size_t was later. I've
2028        moved the #includes for the C headers earlier to avoid this.
2029    
2030    10. Added some adjustments to the code to make it easier to compiler on certain
2031        special systems:
2032    
2033          (a) Some "const" qualifiers were missing.
2034          (b) Added the macro EXPORT before all exported functions; by default this
2035              is defined to be empty.
2036          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2037              that it reads its output file name as an argument instead of writing
2038              to the standard output and assuming this can be redirected.
2039    
2040    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2041        class containing characters with values greater than 255, PCRE compilation
2042        went into a loop.
2043    
2044    12. A recursive reference to a subpattern that was within another subpattern
2045        that had a minimum quantifier of zero caused PCRE to crash. For example,
2046        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2047        recursion. If the recursively-called subpattern itself had a zero repeat,
2048        that was OK.
2049    
2050    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2051        buffer into which it was copied (for escape processing) was still set at
2052        1024, so long lines caused crashes.
2053    
2054    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2055        "internal error: code overflow...". This applied to any character class
2056        that was followed by a possessive quantifier.
2057    
2058    15. Modified the Makefile to add libpcre.la as a prerequisite for
2059        libpcreposix.la because I was told this is needed for a parallel build to
2060        work.
2061    
2062    16. If a pattern that contained .* following optional items at the start was
2063        studied, the wrong optimizing data was generated, leading to matching
2064        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2065        matching string must start with a or b or c. The correct conclusion for
2066        this pattern is that a match can start with any character.
2067    
2068    
2069    Version 4.4 13-Aug-03
2070    ---------------------
2071    
2072     1. In UTF-8 mode, a character class containing characters with values between
2073        127 and 255 was not handled correctly if the compiled pattern was studied.
2074        In fixing this, I have also improved the studying algorithm for such
2075        classes (slightly).
2076    
2077     2. Three internal functions had redundant arguments passed to them. Removal
2078        might give a very teeny performance improvement.
2079    
2080     3. Documentation bug: the value of the capture_top field in a callout is *one
2081        more than* the number of the hightest numbered captured substring.
2082    
2083     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2084        in incorrectly linking with a previously installed version. They now link
2085        explicitly with libpcre.la.
2086    
2087     5. configure.in no longer needs to recognize Cygwin specially.
2088    
2089     6. A problem in pcre.in for Windows platforms is fixed.
2090    
2091     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2092        pcretest, it used to include the size of the study block as part of its
2093        output. Unfortunately, the structure contains a field that has a different
2094        size on different hardware architectures. This meant that the tests that
2095        showed this size failed. As the block is currently always of a fixed size,
2096        this information isn't actually particularly useful in pcretest output, so
2097        I have just removed it.
2098    
2099     8. Three pre-processor statements accidentally did not start in column 1.
2100        Sadly, there are *still* compilers around that complain, even though
2101        standard C has not required this for well over a decade. Sigh.
2102    
2103     9. In pcretest, the code for checking callouts passed small integers in the
2104        callout_data field, which is a void * field. However, some picky compilers
2105        complained about the casts involved for this on 64-bit systems. Now
2106        pcretest passes the address of the small integer instead, which should get
2107        rid of the warnings.
2108    
2109    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2110        both compile and run time, and gives an error if an invalid UTF-8 sequence
2111        is found. There is a option for disabling this check in cases where the
2112        string is known to be correct and/or the maximum performance is wanted.
2113    
2114    11. In response to a bug report, I changed one line in Makefile.in from
2115    
2116            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2117        to
2118            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2119    
2120        to look similar to other lines, but I have no way of telling whether this
2121        is the right thing to do, as I do not use Windows. No doubt I'll get told
2122        if it's wrong...
2123    
2124    
2125    Version 4.3 21-May-03
2126    ---------------------
2127    
2128    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2129       Makefile.
2130    
2131    2. Some refactoring to improve the quality of the code:
2132    
2133       (i)   The utf8_table... variables are now declared "const".
2134    
2135       (ii)  The code for \cx, which used the "case flipping" table to upper case
2136             lower case letters, now just substracts 32. This is ASCII-specific,
2137             but the whole concept of \cx is ASCII-specific, so it seems
2138             reasonable.
2139    
2140       (iii) PCRE was using its character types table to recognize decimal and
2141             hexadecimal digits in the pattern. This is silly, because it handles
2142             only 0-9, a-f, and A-F, but the character types table is locale-
2143             specific, which means strange things might happen. A private
2144             table is now used for this - though it costs 256 bytes, a table is
2145             much faster than multiple explicit tests. Of course, the standard
2146             character types table is still used for matching digits in subject
2147             strings against \d.
2148    
2149       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2150             ending in _t are). So I've renamed it as ESC_tee.
2151    
2152    3. The first argument for regexec() in the POSIX wrapper should have been
2153       defined as "const".
2154    
2155    4. Changed pcretest to use malloc() for its buffers so that they can be
2156       Electric Fenced for debugging.
2157    
2158    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2159       to read one or more bytes before the start of the subject string. Often this
2160       had no effect on PCRE's behaviour, but in some circumstances it could
2161       provoke a segmentation fault.
2162    
2163    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2164       to try to read one or more bytes before the start of the subject string.
2165    
2166    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2167       UTF-8 support could misbehave in various ways if the subject string
2168       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2169       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2170       back over UTF-8 characters.)
2171    
2172    
2173    Version 4.2 14-Apr-03
2174    ---------------------
2175    
2176    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2177    
2178    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2179         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2180         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2181         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2182         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2183           and BUILD_EXEEXT
2184         Note: automatic setting of the BUILD variables is not yet working
2185         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2186           compile-time but not at link-time
2187         [LINK]: use for linking executables only
2188         make different versions for Windows and non-Windows
2189         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2190           libraries
2191         [LINK_FOR_BUILD]: new variable
2192         [OBJEXT]: use throughout
2193         [EXEEXT]: use throughout
2194         <winshared>: new target
2195         <wininstall>: new target
2196         <dftables.o>: use native compiler
2197         <dftables>: use native linker
2198         <install>: handle Windows platform correctly
2199         <clean>: ditto
2200         <check>: ditto
2201         copy DLL to top builddir before testing
2202    
2203       As part of these changes, -no-undefined was removed again. This was reported
2204       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2205       in any case.
2206    
2207    3. Some tidies to get rid of compiler warnings:
2208    
2209       . In the match_data structure, match_limit was an unsigned long int, whereas
2210         match_call_count was an int. I've made them both unsigned long ints.
2211    
2212       . In pcretest the fact that a const uschar * doesn't automatically cast to
2213         a void * provoked a warning.
2214    
2215       . Turning on some more compiler warnings threw up some "shadow" variables
2216         and a few more missing casts.
2217    
2218    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2219       option, a class that contained a single character with a value between 128
2220       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2221    
2222    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2223       option, a class that contained several characters, but with at least one
2224       whose value was between 128 and 255 caused PCRE to crash.
2225    
2226    
2227    Version 4.1 12-Mar-03
2228    ---------------------
2229    
2230    1. Compiling with gcc -pedantic found a couple of places where casts were
2231    needed, and a string in dftables.c that was longer than standard compilers are
2232    required to support.
2233    
2234    2. Compiling with Sun's compiler found a few more places where the code could
2235    be tidied up in order to avoid warnings.
2236    
2237    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2238    first of these names is deprecated in the latest Autoconf in favour of the name
2239    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2240    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2241    analogy I have changed it to CFLAGS_FOR_BUILD.
2242    
2243    4. Added -no-undefined to the linking command in the Makefile, because this is
2244    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2245    linking step for the pcreposix library.
2246    
2247    5. PCRE was failing to diagnose the case of two named groups with the same
2248    name.
2249    
2250    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2251    literal character that is needed in the subject for a match, and scans along to
2252    ensure that it is present before embarking on the full matching process. This
2253    saves time in cases of nested unlimited repeats that are never going to match.
2254    Problem: the scan can take a lot of time if the subject is very long (e.g.
2255    megabytes), thus penalizing straightforward matches. It is now done only if the
2256    amount of subject to be scanned is less than 1000 bytes.
2257    
2258    7. A lesser problem with the same optimization is that it was recording the
2259    first character of an anchored pattern as "needed", thus provoking a search
2260    right along the subject, even when the first match of the pattern was going to
2261    fail. The "needed" character is now not set for anchored patterns, unless it
2262    follows something in the pattern that is of non-fixed length. Thus, it still
2263    fulfils its original purpose of finding quick non-matches in cases of nested
2264    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2265    
2266    
2267    Version 4.0 17-Feb-03
2268    ---------------------
2269    
2270    1. If a comment in an extended regex that started immediately after a meta-item
2271    extended to the end of string, PCRE compiled incorrect data. This could lead to
2272    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2273    
2274    2. Moved to autoconf 2.53 and libtool 1.4.2.
2275    
2276    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2277    the special perltest8 script is no longer needed - all the tests can be run
2278    from a single perltest script.
2279    
2280    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2281    by \s. It has now been removed in PCRE. This means it isn't recognized as
2282    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2283    class [:space:] *does* include VT, thereby creating a mess.
2284    
2285    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2286    space and tab.
2287    
2288    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2289    its new features into the main test script, reducing the number of scripts.
2290    
2291    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2292    were backward compatible, and made the (?i) apply to the whole pattern, as if
2293    /i were given. Now it behaves more logically, and applies the option setting
2294    only to what follows. PCRE has been changed to follow suit. However, if it
2295    finds options settings right at the start of the pattern, it extracts them into
2296    the global options, as before. Thus, they show up in the info data.
2297    
2298    8. Added support for the \Q...\E escape sequence. Characters in between are
2299    treated as literals. This is slightly different from Perl in that $ and @ are
2300    also handled as literals inside the quotes. In Perl, they will cause variable
2301    interpolation. Note the following examples:
2302    
2303        Pattern            PCRE matches      Perl matches
2304    
2305        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2306        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2307        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2308    
2309    For compatibility with Perl, \Q...\E sequences are recognized inside character
2310    classes as well as outside them.
2311    
2312    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2313    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2314    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2315    signed/unsigned warnings.
2316    
2317    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2318    option for pcretest, so I've replaced it by a simple function that does just
2319    that job.
2320    
2321    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2322    "pcregrep -".
2323    
2324    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2325    Java package. This provides some syntactic sugar for simple cases of what my
2326    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2327    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2328    item, you can use this simplified notation. Note that only makes sense with
2329    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2330    greediness, whatever the setting of the PCRE_UNGREEDY option.
2331    
2332    13. A change of greediness default within a pattern was not taking effect at
2333    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2334    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2335    was abstracted outside.
2336    
2337    14. PCRE now supports the \G assertion. It is true when the current matching
2338    position is at the start point of the match. This differs from \A when the
2339    starting offset is non-zero. Used with the /g option of pcretest (or similar
2340    code), it works in the same way as it does for Perl's /g option. If all
2341    alternatives of a regex begin with \G, the expression is anchored to the start
2342    match position, and the "anchored" flag is set in the compiled expression.
2343    
2344    15. Some bugs concerning the handling of certain option changes within patterns
2345    have been fixed. These applied to options other than (?ims). For example,
2346    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2347    been the other way round. Some of this was related to change 7 above.
2348    
2349    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2350    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2351    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2352    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2353    
2354    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2355    mode. Unlike ".", it always matches newline, whatever the setting of
2356    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2357    assertions. Perl allows it, but it doesn't (in general) work because it can't
2358    calculate the length of the lookbehind. At least, that's the case for Perl
2359    5.8.0 - I've been told they are going to document that it doesn't work in
2360    future.
2361    
2362    18. Added an error diagnosis for escapes that PCRE does not support: these are
2363    \L, \l, \N, \P, \p, \U, \u, and \X.
2364    
2365    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2366    reading past the end of the pattern in cases such as /[abcd/.
2367    
2368    20. PCRE was getting more memory than necessary for patterns with classes that
2369    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2370    
2371    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2372    compiling PCRE for use with Virtual Pascal.
2373    
2374    22. Small fix to the Makefile to make it work properly if the build is done
2375    outside the source tree.
2376    
2377    23. Added a new extension: a condition to go with recursion. If a conditional
2378    subpattern starts with (?(R) the "true" branch is used if recursion has
2379    happened, whereas the "false" branch is used only at the top level.
2380    
2381    24. When there was a very long string of literal characters (over 255 bytes
2382    without UTF support, over 250 bytes with UTF support), the computation of how
2383    much memory was required could be incorrect, leading to segfaults or other
2384    strange effects.
2385    
2386    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2387    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2388    there was a subsequent back reference to those brackets. This meant that, for
2389    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2390    possible to check for precisely this case. All we can do is abandon the
2391    optimization if .* occurs inside capturing brackets when there are any back
2392    references whatsoever. (See below for a better fix that came later.)
2393    
2394    26. The handling of the optimization for finding the first character of a
2395    non-anchored pattern, and for finding a character that is required later in the
2396    match were failing in some cases. This didn't break the matching; it just
2397    failed to optimize when it could. The way this is done has been re-implemented.
2398    
2399    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2400    
2401    28. Added a new feature that provides some of the functionality that Perl
2402    provides with (?{...}). The facility is termed a "callout". The way it is done
2403    in PCRE is for the caller to provide an optional function, by setting
2404    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2405    global variable. By default it is unset, which disables all calling out. To get
2406    the function called, the regex must include (?C) at appropriate points. This
2407    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2408    This provides a means of identifying different callout points. When PCRE
2409    reaches such a point in the regex, if pcre_callout has been set, the external
2410    function is called. It is provided with data in a structure called
2411    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2412    matching continues; if it returns a non-zero value, the match at the current
2413    point fails. However, backtracking will occur if possible. [This was changed
2414    later and other features added - see item 49 below.]
2415    
2416    29. pcretest is upgraded to test the callout functionality. It provides a
2417    callout function that displays information. By default, it shows the start of
2418    the match and the current position in the text. There are some new data escapes
2419    to vary what happens:
2420    
2421        \C+         in addition, show current contents of captured substrings
2422        \C-         do not supply a callout function
2423        \C!n        return 1 when callout number n is reached
2424        \C!n!m      return 1 when callout number n is reached for the mth time
2425    
2426    30. If pcregrep was called with the -l option and just a single file name, it
2427    output "<stdin>" if a match was found, instead of the file name.
2428    
2429    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2430    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2431    pcre_exec(). This saves a malloc/free per call. The default value of
2432    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2433    when configuring.
2434    
2435    32. The default maximum size of a compiled pattern is 64K. There have been a
2436    few cases of people hitting this limit. The code now uses macros to handle the
2437    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2438    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2439    configuring. Tests 2 and 5 work only with 2-byte links because they output
2440    debugging information about compiled patterns.
2441    
2442    33. Internal code re-arrangements:
2443    
2444    (a) Moved the debugging function for printing out a compiled regex into
2445        its own source file (printint.c) and used #include to pull it into
2446        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2447        separate copies.
2448    
2449    (b) Defined the list of op-code names for debugging as a macro in
2450        internal.h so that it is next to the definition of the opcodes.
2451    
2452    (c) Defined a table of op-code lengths for simpler skipping along compiled
2453        code. This is again a macro in internal.h so that it is next to the
2454        definition of the opcodes.
2455    
2456    34. Added support for recursive calls to individual subpatterns, along the
2457    lines of Robin Houston's patch (but implemented somewhat differently).
2458    
2459    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2460    allow it to read and process whole directories in Win32. This code was
2461    contributed by Lionel Fourquaux; it has not been tested by me.
2462    
2463    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2464    used to name a group. Names consist of alphanumerics and underscores, and must
2465    be unique. Back references use the syntax (?P=name) and recursive calls use
2466    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2467    numbers. The function pcre_fullinfo() can be used after compilation to extract
2468    a name/number map. There are three relevant calls:
2469    
2470      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2471      PCRE_INFO_NAMECOUNT            yields the number of entries
2472      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2473    
2474    The map is a vector of fixed-size entries. The size of each entry depends on
2475    the length of the longest name used. The first two bytes of each entry are the
2476    group number, most significant byte first. There follows the corresponding
2477    name, zero terminated. The names are in alphabetical order.
2478    
2479    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2480    case instead of 255. Making it the same both with and without UTF-8 support
2481    means that the same test output works with both.
2482    
2483    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2484    calling malloc() with a zero argument.
2485    
2486    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2487    optimization. I've improved things by keeping a bitmap of backreferences with
2488    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2489    fact referenced, the optimization can be applied. It is unlikely that a
2490    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2491    the match to follow \n) will appear inside brackets with a number greater than
2492    31, but if it does, any back reference > 31 suppresses the optimization.
2493    
2494    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2495    of disabling numbered capturing parentheses. Any opening parenthesis that is
2496    not followed by ? behaves as if it were followed by ?: but named parentheses
2497    can still be used for capturing (and they will acquire numbers in the usual
2498    way).
2499    
2500    41. Redesigned the return codes from the match() function into yes/no/error so
2501    that errors can be passed back from deep inside the nested calls. A malloc
2502    failure while inside a recursive subpattern call now causes the
2503    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2504    
2505    42. It is now possible to set a limit on the number of times the match()
2506    function is called in a call to pcre_exec(). This facility makes it possible to
2507    limit the amount of recursion and backtracking, though not in a directly
2508    obvious way, because the match() function is used in a number of different
2509    circumstances. The count starts from zero for each position in the subject
2510    string (for non-anchored patterns). The default limit is, for compatibility, a
2511    large number, namely 10 000 000. You can change this in two ways:
2512    
2513    (a) When configuring PCRE before making, you can use --with-match-limit=n
2514        to set a default value for the compiled library.
2515    
2516    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2517        a different value is set. See 45 below.
2518    
2519    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2520    
2521    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2522    of things that can be changed at compile time. The first argument specifies
2523    what is wanted and the second points to where the information is to be placed.
2524    The current list of available information is:
2525    
2526      PCRE_CONFIG_UTF8
2527    
2528    The output is an integer that is set to one if UTF-8 support is available;
2529    otherwise it is set to zero.
2530    
2531      PCRE_CONFIG_NEWLINE
2532    
2533    The output is an integer that it set to the value of the code that is used for
2534    newline. It is either LF (10) or CR (13).
2535    
2536      PCRE_CONFIG_LINK_SIZE
2537    
2538    The output is an integer that contains the number of bytes used for internal
2539    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2540    
2541      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2542    
2543    The output is an integer that contains the threshold above which the POSIX
2544    interface uses malloc() for output vectors. See item 31 above.
2545    
2546      PCRE_CONFIG_MATCH_LIMIT
2547    
2548    The output is an unsigned integer that contains the default limit of the number
2549    of match() calls in a pcre_exec() execution. See 42 above.
2550    
2551    44. pcretest has been upgraded by the addition of the -C option. This causes it
2552    to extract all the available output from the new pcre_config() function, and to
2553    output it. The program then exits immediately.
2554    
2555    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2556    order to support additional features. One way would have been to define
2557    pcre_exec2() (for example) with extra arguments, but this would not have been
2558    extensible, and would also have required all calls to the original function to
2559    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2560    is used for passing in "extra" data from pcre_study().
2561    
2562    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2563    contains the following fields:
2564    
2565      flags         a bitmap indicating which of the following fields are set
2566      study_data    opaque data from pcre_study()
2567      match_limit   a way of specifying a limit on match() calls for a specific
2568                      call to pcre_exec()
2569      callout_data  data for callouts (see 49 below)
2570    
2571    The flag bits are also defined in pcre.h, and are
2572    
2573      PCRE_EXTRA_STUDY_DATA
2574      PCRE_EXTRA_MATCH_LIMIT
2575      PCRE_EXTRA_CALLOUT_DATA
2576    
2577    The pcre_study() function now returns one of these new pcre_extra blocks, with
2578    the actual study data pointed to by the study_data field, and the
2579    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2580    before. That is, this change is entirely upwards-compatible and requires no
2581    change to existing code.
2582    
2583    If you want to pass in additional data to pcre_exec(), you can either place it
2584    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2585    block.
2586    
2587    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2588    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2589    times with different match limits, until it finds the minimum value needed for
2590    pcre_exec() to complete. The value is then output. This can be instructive; for
2591    most simple matches the number is quite small, but for pathological cases it
2592    gets very large very quickly.
2593    
2594    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2595    returns the size of the data block pointed to by the study_data field in a
2596    pcre_extra block, that is, the value that was passed as the argument to
2597    pcre_malloc() when PCRE was getting memory in which to place the information
2598    created by pcre_study(). The fourth argument should point to a size_t variable.
2599    pcretest has been extended so that this information is shown after a successful
2600    pcre_study() call when information about the compiled regex is being displayed.
2601    
2602    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2603    because what follows is always an absolute path. (Later: it turns out that this
2604    is more than cosmetic for MinGW, because it doesn't like empty path
2605    components.)
2606    
2607    49. Some changes have been made to the callout feature (see 28 above):
2608    
2609    (i)  A callout function now has three choices for what it returns:
2610    
2611           0  =>  success, carry on matching
2612         > 0  =>  failure at this point, but backtrack if possible
2613         < 0  =>  serious error, return this value from pcre_exec()
2614    
2615         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2616         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2617         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2618         use by callout functions. It will never be used by PCRE itself.
2619    
2620    (ii) The pcre_extra structure (see 45 above) has a void * field called
2621         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2622         pcre_callout_block structure has a field of the same name. The contents of
2623         the field passed in the pcre_extra structure are passed to the callout
2624         function in the corresponding field in the callout block. This makes it
2625         easier to use the same callout-containing regex from multiple threads. For
2626         testing, the pcretest program has a new data escape
2627    
2628           \C*n        pass the number n (may be negative) as callout_data
2629    
2630         If the callout function in pcretest receives a non-zero value as
2631         callout_data, it returns that value.
2632    
2633    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2634    there were some redundant $(CFLAGS) in commands that are now specified as
2635    $(LINK), which already includes $(CFLAGS).
2636    
2637    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2638    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2639    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2640    one-byte characters throughout. Note that case-insensitive matching applies
2641    only to characters whose values are less than 256. PCRE doesn't support the
2642    notion of cases for higher-valued characters.
2643    
2644    (i)   A character class whose characters are all within 0-255 is handled as
2645          a bit map, and the map is inverted for negative classes. Previously, a
2646          character > 255 always failed to match such a class; however it should
2647          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2648    
2649    (ii)  A negated character class with a single character < 255 is coded as
2650          "not this character" (OP_NOT). This wasn't working properly when the test
2651          character was multibyte, either singly or repeated.
2652    
2653    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2654          mode, for example: \x{100}{2,3}.
2655    
2656    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2657          singly or repeated) now correctly test multibyte characters. However,
2658          PCRE doesn't recognize any characters with values greater than 255 as
2659          digits, spaces, or word characters. Such characters always match \D, \S,
2660          and \W, and never match \d, \s, or \w.
2661    
2662    (v)   Classes may now contain characters and character ranges with values
2663          greater than 255. For example: [ab\x{100}-\x{400}].
2664    
2665    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2666          PCRE in UTF-8 mode.
2667    
2668    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2669    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2670    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2671    value.)
2672    
2673    53. The single man page has become too large. I have therefore split it up into
2674    a number of separate man pages. These also give rise to individual HTML pages;
2675    these are now put in a separate directory, and there is an index.html page that
2676    lists them all. Some hyperlinking between the pages has been installed.
2677    
2678    54. Added convenience functions for handling named capturing parentheses.
2679    
2680    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2681    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2682    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2683    are faulted.
2684    
2685    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2686    calling configure. These values are used when compiling the dftables.c program
2687    which is run to generate the source of the default character tables. They
2688    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2689    you will need to set these values.
2690    
2691    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2692    
2693    
2694    Version 3.9 02-Jan-02
2695    ---------------------
2696    
2697    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2698    
2699    2. If --disable-static was given, the building process failed when trying to
2700    build pcretest and pcregrep. (For some reason it was using libtool to compile
2701    them, which is not right, as they aren't part of the library.)
2702    
2703    
2704    Version 3.8 18-Dec-01
2705    ---------------------
2706    
2707    1. The experimental UTF-8 code was completely screwed up. It was packing the
2708    bytes in the wrong order. How dumb can you get?
2709    
2710    
2711    Version 3.7 29-Oct-01
2712    ---------------------
2713    
2714    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2715    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2716    this didn't happen under Solaris 8, where I normally test things.
2717    
2718    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2719    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2720    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2721    this fix an hour or so after the initial 3.7 release.)
2722    
2723    
2724    Version 3.6 23-Oct-01
2725    ---------------------
2726    
2727    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2728    offsets passed as NULL with zero offset count.
2729    
2730    2. The config.guess and config.sub files had not been updated when I moved to
2731    the latest autoconf.
2732    
2733    
2734    Version 3.5 15-Aug-01
2735    ---------------------
2736    
2737    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2738    had been forgotten.
2739    
2740    2. By using declared but undefined structures, we can avoid using "void"
2741    definitions in pcre.h while keeping the internal definitions of the structures
2742    private.
2743    
2744    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2745    user point of view, this means that both static and shared libraries are built
2746    by default, but this can be individually controlled. More of the work of
2747    handling this static/shared cases is now inside libtool instead of PCRE's make
2748    file.
2749    
2750    4. The pcretest utility is now installed along with pcregrep because it is
2751    useful for users (to test regexs) and by doing this, it automatically gets
2752    relinked by libtool. The documentation has been turned into a man page, so
2753    there are now .1, .txt, and .html versions in /doc.
2754    
2755    5. Upgrades to pcregrep:
2756       (i)   Added long-form option names like gnu grep.
2757       (ii)  Added --help to list all options with an explanatory phrase.
2758       (iii) Added -r, --recursive to recurse into sub-directories.
2759       (iv)  Added -f, --file to read patterns from a file.
2760    
2761    6. pcre_exec() was referring to its "code" argument before testing that
2762    argument for NULL (and giving an error if it was NULL).
2763    
2764    7. Upgraded Makefile.in to allow for compiling in a different directory from
2765    the source directory.
2766    
2767    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2768    options bits, the pointer it was passed was to an int instead of to an unsigned
2769    long int. This mattered only on 64-bit systems.
2770    
2771    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2772    generated) instead of pcre.in, which it its source. Also made the same change
2773    in several of the .c files.
2774    
2775    10. A new release of gcc defines printf() as a macro, which broke pcretest
2776    because it had an ifdef in the middle of a string argument for printf(). Fixed
2777    by using separate calls to printf().
2778    
2779    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2780    script, to force use of CR or LF instead of \n in the source. On non-Unix
2781    systems, the value can be set in config.h.
2782    
2783    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2784    absolute limit. Changed the text of the error message to make this clear, and
2785    likewise updated the man page.
2786    
2787    13. The limit of 99 on the number of capturing subpatterns has been removed.
2788    The new limit is 65535, which I hope will not be a "real" limit.
2789    
2790    
2791    Version 3.4 22-Aug-00
2792    ---------------------
2793    
2794    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2795    
2796    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2797    
2798    
2799    Version 3.3 01-Aug-00
2800    ---------------------
2801    
2802    1. If an octal character was given, but the value was greater than \377, it
2803    was not getting masked to the least significant bits, as documented. This could
2804    lead to crashes in some systems.
2805    
2806    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2807    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2808    
2809    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2810    These just pass their arguments on to (pcre_free)(), but they are provided
2811    because some uses of PCRE bind it to non-C systems that can call its functions,
2812    but cannot call free() or pcre_free() directly.
2813    
2814    4. Add "make test" as a synonym for "make check". Corrected some comments in
2815    the Makefile.
2816    
2817    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2818    Makefile.
2819    
2820    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2821    command called pgrep for grepping around the active processes.
2822    
2823    7. Added the beginnings of support for UTF-8 character strings.
2824    
2825    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2826    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2827    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2828    out for the ar command.)
2829    
2830    
2831    Version 3.2 12-May-00
2832    ---------------------
2833    
2834    This is purely a bug fixing release.
2835    
2836    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2837    of ZA. This was just one example of several cases that could provoke this bug,
2838    which was introduced by change 9 of version 2.00. The code for breaking
2839    infinite loops after an iteration that matches an empty string was't working
2840    correctly.
2841    
2842    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2843    when matched against abbab (for example). After matching an empty string, it
2844    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2845    caused it to match further down the string than it should.
2846    
2847    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2848    was there because it doesn't seem to be needed, and it causes trouble on some
2849    systems, as it is not a Standard C header. It has been removed.
2850    
2851    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2852    were reported on the Macintosh. The changes were from
2853    
2854      while ((c = *(++ptr)) != 0 && c != '\n');
2855    to
2856      while ((c = *(++ptr)) != 0 && c != '\n') ;
2857    
2858    Totally extraordinary, but if that's what it takes...
2859    
2860    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2861    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2862    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2863    assumes the way PCRE uses memmove() (always moving upwards).
2864    
2865    6. PCRE is being used in one environment where strchr() is not available. There
2866    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2867    faster code anyway.
2868    
2869    
2870    Version 3.1 09-Feb-00
2871    ---------------------
2872    
2873    The only change in this release is the fixing of some bugs in Makefile.in for
2874    the "install" target:
2875    
2876    (1) It was failing to install pcreposix.h.
2877    
2878    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2879    
2880    
2881    Version 3.0 01-Feb-00
2882    ---------------------
2883    
2884    1. Add support for the /+ modifier to perltest (to output $` like it does in
2885    pcretest).
2886    
2887    2. Add support for the /g modifier to perltest.
2888    
2889    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2890    matches null strings.
2891    
2892    4. Fix perltest so that it doesn't do unwanted things when fed an empty
2893    pattern. Perl treats empty patterns specially - it reuses the most recent
2894    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2895    effect.
2896    
2897    5. The POSIX interface was broken in that it was just handing over the POSIX
2898    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2899    required a bigger vector, with some working space on the end. This means that
2900    the POSIX wrapper now has to get and free some memory, and copy the results.
2901    
2902    6. Added some simple autoconf support, placing the test data and the
2903    documentation in separate directories, re-organizing some of the
2904    information files, and making it build pcre-config (a GNU standard). Also added
2905    libtool support for building PCRE as a shared library, which is now the
2906    default.
2907    
2908    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2909    09 are not valid octal constants. Single digits will be used for minor values
2910    less than 10.
2911    
2912    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2913    existing programs that set these in the POSIX interface can use PCRE without
2914    modification.
2915    
2916    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2917    return all that pcre_info() returns, plus additional data. The pcre_info()
2918    function is retained for compatibility, but is considered to be obsolete.
2919    
2920    10. Added experimental recursion feature (?R) to handle one common case that
2921    Perl 5.6 will be able to do with (?p{...}).
2922    
2923    11. Added support for POSIX character classes like [:alpha:], which Perl is
2924    adopting.
2925    
2926    
2927    Version 2.08 31-Aug-99
2928    ----------------------
2929    
2930    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2931    trying to match at the startoffset position, but instead was moving forward to
2932    the next newline as if a previous match had failed.
2933    
2934    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2935    and could get into a loop if a null string was matched other than at the start
2936    of the subject.
2937    
2938    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2939    be distinguished at compile time, and for completeness also added PCRE_DATE.
2940    
2941    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2942    in GnuWin32 environments.
2943    
2944    
2945    Version 2.07 29-Jul-99
2946    ----------------------
2947    
2948    1. The documentation is now supplied in plain text form and HTML as well as in
2949    the form of man page sources.
2950    
2951    2. C++ compilers don't like assigning (void *) values to other pointer types.
2952    In particular this affects malloc(). Although there is no problem in Standard
2953    C, I've put in casts to keep C++ compilers happy.
2954    
2955    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2956    should be (const char *).
2957    
2958    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2959    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2960    However, I haven't made this a standard facility. The documentation doesn't
2961    mention it, and the Makefile doesn't support it.
2962    
2963    5. The Makefile now contains an "install" target, with editable destinations at
2964    the top of the file. The pcretest program is not installed.
2965    
2966    6. pgrep -V now gives the PCRE version number and date.
2967    
2968    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2969    causing the entire string to be ignored, instead of just the last character.
2970    
2971    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2972    non-matching string, it can take a very, very long time, even for strings of
2973    quite modest length, because of the nested recursion. PCRE now does better in
2974    some of these cases. It does this by remembering the last required literal
2975    character in the pattern, and pre-searching the subject to ensure it is present
2976    before running the real match. In other words, it applies a heuristic to detect
2977    some types of certain failure quickly, and in the above example, if presented
2978    with a string that has no trailing " it gives "no match" very quickly.
2979    
2980    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2981    other alternatives are tried instead.
2982    
2983    
2984    Version 2.06 09-Jun-99
2985    ----------------------
2986    
2987    1. Change pcretest's output for amount of store used to show just the code
2988    space, because the remainder (the data block) varies in size between 32-bit and
2989    64-bit systems.
2990    
2991    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2992    start matching at. This allows lookbehinds to work when searching for multiple
2993    occurrences in a string.
2994    
2995    3. Added additional options to pcretest for testing multiple occurrences:
2996    
2997       /+   outputs the rest of the string that follows a match
2998       /g   loops for multiple occurrences, using the new startoffset argument
2999       /G   loops for multiple occurrences by passing an incremented pointer
3000    
3001    4. PCRE wasn't doing the "first character" optimization for patterns starting
3002    with \b or \B, though it was doing it for other lookbehind assertions. That is,
3003    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3004    the letter 'x'. On long subject strings, this gives a significant speed-up.
3005    
3006    
3007    Version 2.05 21-Apr-99
3008    ----------------------
3009    
3010    1. Changed the type of magic_number from int to long int so that it works
3011    properly on 16-bit systems.
3012    
3013    2. Fixed a bug which caused patterns starting with .* not to work correctly
3014    when the subject string contained newline characters. PCRE was assuming
3015    anchoring for such patterns in all cases, which is not correct because .* will
3016    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3017    DOTALL is set at top level; otherwise it knows that patterns starting with .*
3018    must be retried after every newline in the subject.
3019    
3020    
3021    Version 2.04 18-Feb-99
3022    ----------------------
3023    
3024    1. For parenthesized subpatterns with repeats whose minimum was zero, the
3025    computation of the store needed to hold the pattern was incorrect (too large).
3026    If such patterns were nested a few deep, this could multiply and become a real
3027    problem.
3028    
3029    2. Added /M option to pcretest to show the memory requirement of a specific
3030    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3031    
3032    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3033    compiled in such a way that the backtracking after subsequent failure was
3034    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3035    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3036    
3037    
3038    Version 2.03 02-Feb-99
3039    ----------------------
3040    
3041    1. Fixed typo and small mistake in man page.
3042    
3043    2. Added 4th condition (GPL supersedes if conflict) and created separate
3044    LICENCE file containing the conditions.
3045    
3046    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3047    Perl, that is the internal \ allows the delimiter to be included in the
3048    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3049    the final delimiter, add \ to the end of the pattern (to test the error).
3050    
3051    4. Added the convenience functions for extracting substrings after a successful
3052    match. Updated pcretest to make it able to test these functions.
3053    
3054    
3055    Version 2.02 14-Jan-99
3056    ----------------------
3057    
3058    1. Initialized the working variables associated with each extraction so that
3059    their saving and restoring doesn't refer to uninitialized store.
3060    
3061    2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3062    compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3063    fix the problem.
3064    
3065    3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3066    calls, and wasn't printing the correct value for compiling calls. Increased the
3067    default value of LOOPREPEAT, and the number of significant figures in the
3068    times.
3069    
3070    4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3071    
3072    5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3073    a building problem on Windows NT with a FAT file system.
3074    
3075    
3076  Version 2.01 21-Oct-98  Version 2.01 21-Oct-98
3077  ----------------------  ----------------------

Legend:
Removed from v.25  
changed lines
  Added in v.388

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12